OPTIMISING REOVIRUS TYPE 3 DEARING (REOLYSIN®) AS AN …epubs.surrey.ac.uk/811663/1/FINAL THESIS...

I

OPTIMISING REOVIRUS TYPE 3

DEARING (REOLYSIN®) AS AN ANTI-

CANCER THERAPEUTIC

Gemma Bolton

A thesis submitted in accordance with the requirements of the

University of Surrey for the degree of Doctor of Philosophy

Faculty of Health and Medical Sciences

Department of Microbial and Cellular Sciences

University of Surrey

May 2016

II

DECLARATION OF ORIGINALITY

This thesis and the work to which it refers are the results of my own efforts. Any

ideas, data, images or text resulting from the work of others (whether published or

unpublished) are fully identified as such within the work and attributed to their

originator in the text, bibliography or in footnotes. This thesis has not been submitted

in whole or in part for any other academic degree or professional qualification. I

agree that the University has the right to submit my work to the plagiarism detection

service TurnitinUK for originality checks. Whether or not drafts have been so-

assessed, the University reserves the right to require an electronic version of the final

document (as submitted) for assessment as above.

Gemma Bolton

May 2016

III

ABSTRACT

Reolysin® is a naturally occurring, replication competent formulation of reovirus

Type 3 Dearing (T3D), which has displayed oncolytic activity in a variety of human

cancers in vitro and in clinical trials. Reolysin® shows great promise as a cancer

therapeutic, but further optimisation is needed to maximise its oncolytic potential.

Previous work has failed to uncover the full mechanism of reovirus oncolysis, and this

has hindered the discovery of a reliable biomarker of reovirus treatment response.

Through gene expression profiling, knock-down, and over-expression experiments,

we have identified Yes-Associated Protein-1 (YAP1) as a host-cell factor that predicts

the susceptibility of squamous cell carcinoma of the head and neck (SCCHN) cell

lines to reovirus-induced cell death. YAP1 is a downstream effector of the Hippo

pathway that regulates cellular growth and, sometimes, cancer progression.

Mechanistic studies revealed that YAP1-mediated restriction of reovirus oncolysis

may partially affect direct reovirus replication, but does not occur at the cell surface

via the main reovirus receptor, JAM-A, nor is it linked to the type I interferon anti-

viral response. YAP1 protein expression appears to be cancer-specific; 13% of head

and neck carcinoma tissues stained positive for YAP1, but expression was negligible

in tissue derived from normal organs. Therefore, YAP1 shows potential as a clinical

biomarker to help characterise the most responsive SCCHN patient subgroup to

reovirus treatment, which warrants further investigation.

Additionally, combining reovirus with 3 weekly chemotherapy regimens at the

standard maximum tolerated dose (MTD) may not be the optimum mode of

administration, as the drug-free breaks often allow tumour-vasculature re-growth,

resulting in disease progression. Metronomic chemotherapy (MC) primarily targets

endothelial cells that support tumour-associated angiogenesis, and is less toxic than

the MTD. We have investigated a novel treatment combination of reovirus and low

doses of taxane chemotherapy agents in prostate cancer (PCa) cell lines. The

interaction of reovirus and Cabazitaxel or Docetaxel at doses considerably less than

their IC50 values was measured by using cell viability assays and the Bliss statistical

model, which demonstrated synergistic anti-cancer activity. This was partly due to

enhanced microtubule stabilisation. Our data provides substance to assess the

efficacy of this type of combination therapy in vivo, and subsequently in human trials.

IV

ACKNOWLEDGEMENTS

I would like to express my gratitude to my supervisors, Prof Hardev Pandha and Dr

Guy Simpson, for their patience, knowledge, motivation and continuous support

throughout this work. I have gained a great deal of experience whilst working with

them, and I am indebted for the time and effort they have invested in me.

I would like to thank members of the Oncology Department, within the Faculty of

Health and Medical Sciences, whose friendly advice has contributed to the work of

this thesis. Many thanks to Dr Nicola Annels for her kind assistance with the

confocal microscope and proof-reading. I thank Oncolytics Biotech for providing me

with the funding to support my research.

Finally, I am forever grateful to my friends and family, particularly my husband and

soul mate Kevin, for the love and support throughout my studies. A big thank you to

my brother Paul, for the encouragement and making the tough times bearable.

This thesis is dedicated to the memory of my dear father, Danny Morgan, whose

words of wisdom were, and continue to be, an inspiration to my life.

V

TABLE OF CONTENTS

DECLARATION OF ORIGINALITY II

ABSTRACT III

ACKNOWLEDGEMENTS IV

TABLE OF CONTENTS V

LIST OF ABBREVIATIONS XII

LIST OF FIGURES XIX

LIST OF TABLES XXIV

CHAPTER 1 1

1. INTRODUCTION 2

1.1. CANCER 2

1.1.1. Hallmarks 2

1.1.2. Head and Neck Cancer 3

1.1.2.1. Incidence 3

1.1.2.2. Risk factors and symptoms 4

1.1.2.3. Heterogeneity 4

1.1.2.4. Pathogenesis 5

1.1.2.5. Diagnosis, treatment and prognosis 5

1.1.3. Prostate Cancer 6

1.1.3.1. Incidence 6

1.1.3.2. Risk factors and symptoms 6

1.1.3.3. Heterogeneity and pathogenesis 7

1.1.3.4. Diagnosis, treatment and prognosis 8

1.2. ONCOLYTIC VIROTHERAPY 9

1.2.1. History 9

1.2.2. Oncolytic Adenovirus 9

1.2.3. Oncolytic Herpes Simplex virus 11

1.2.4. Oncolytic Vaccinia virus 12

1.2.5. Oncolytic Newcastle Disease virus 12

1.2.6. Oncolytic Vesicular Stomatitis virus 13

1.2.7. Oncolytic Coxsackie virus 13

1.3. ONCOLYTIC REOVIRUS 16

1.3.1. dsRNA genome and molecular structure 16

1.3.2. Replication life cycle 17

1.3.3. Mechanism of selective replication in cancer cells 19

VI

1.3.3.1. Targeting of an aberrant Ras signalling

pathway 19

1.3.3.2. Studies that conflict the involvement of

aberrant Ras signalling 21

1.3.4. Biomarkers of treatment response 22

1.3.5. Pre-clinical testing of Oncolytic Reovirus T3D 23

1.3.6. Clinical trials involving Oncolytic Reovirus T3D 24

1.3.7. Combining Reovirus T3D with metronomic doses of

taxane chemotherapy drugs 29

1.4. SUMMARY 33

1.5. HYPOTHESIS AND OBJECTIVES 34

CHAPTER 2 35

2. MATERIALS AND METHODS 36

2.1. Reovirus and Chemotherapeutic taxane drugs 36

2.2. Cell culture media 36

2.3. Cell lines 37

2.4. Passaging of adherent cells 38

2.5. Evaluation of cell number 38

2.6. Cryopreservation of cells 38

2.7. Revitalisation of cryopreserved cells 39

2.8. Calculating the volume of reovirus needed for a certain

multiplicity of infection (MOI) 39

2.9. Cell titre 96® aqueous non-radioactive cell proliferation (MTS)

assay 40

2.10. RNA extraction from cell lines 41

2.11. Complementary DNA (cDNA) synthesis from cell lines 41

2.12. Real time-quantitative polymerase chain reaction (RT-qPCR) 42

2.13. siRNA-mediated gene knock-down in the PJ41 cell line 44

2.13.1 KDAlert™ GAPDH assay kit for detection of GAPDH

knock-down 44

2.13.2. siRNA-mediated knock-down of a target gene 46

2.13.3. siRNA-mediated knock-down of a target gene and

infection with reovirus 47

2.14. Western blotting for protein detection in cell lysates 48

2.14.1. Lysate preparation and protein separation by sodium

dodecyl sulfate polyacrylamide gel electrophoresis

(SDS-PAGE)

48

2.14.2. Protein transfer, blocking, antibody probing and band

detection 48

2.14.3. Stripping the membrane and antibody re-probing 49

2.14.4. Densitometry analysis 49

VII

2.15. Bacterial transformation and purification of plasmid DNA 50

2.15.1. Bacterial transformation and making a glycerol stock 50

2.15.2. Purification of the plasmid DNA 51

2.16. Lipid-mediated over-expression of YAP1 in cell lines 52

2.16.1. Transient over-expression of YAP1 52

2.16.2. Stable over-expression of YAP1 in the HN5 SCCHN

cell line 53

2.16.3. Over-expression of YAP1 and reovirus infection 55

2.17. Detection of a protein using immunofluorescence staining and

confocal microscopy in cell lines 55

2.18. The effect of sphingosine-1-phosphate (S1P) on YAP1 activity

and reovirus oncolysis 57

2.19. Detection of a protein by indirect flow cytometry 58

2.20. One step growth curve analysis by the 50% tissue culture

infective dose (TCID50) assay 59

2.20.1. Preparation of intracellular or extracellular viral

samples 59

2.20.2. Infection of the host-cell monolayer and determining

the cytopathic effect 60

2.21. Verikine™ human interferon-beta (IFN-β) enzyme-linked

immunosorbent assay (ELISA) 60

2.21.1. Isolation of peripheral blood mononuclear cells

(PBMCs) from whole blood 60

2.21.2. Preparation of cell supernatants 61

2.21.3. The Verikine™ Human IFN-β ELISA assay 61

2.22. Detection of the YAP1 protein in tissue by enzymatic

immunohistochemistry (IHC) staining 62

2.22.1. Deparaffinization and antigen retrieval 62

2.22.2. Blocking of the tissue and addition of antibodies 63

2.22.3. Addition of the Avidin-biotin complex (ABC) and

3,3’diaminobenzidine (DAB) substrate 63

2.22.4. Dehydration of the tissue sections, cover-slipping and

scoring 64

2.23. Assessing the interaction between reovirus and taxane

chemotherapeutic drugs in PCa cell lines 65

2.23.1. Concurrent combination of two agents at fixed-dose

ratios 65

2.23.2. Comparing sequential and concurrent combinations at

fixed-dose ratios 65

2.23.3. Concurrent combination of two agents at non-fixed dose

ratios 66

2.24. One step growth curve analysis by the virus plaque assay 67

VIII

2.24.1. Infection of the host-cell monolayer and counting

plaques 67

2.25. Inhibition of apoptosis by z-VAD-FMK 68

2.26. Inhibition of necroptosis by necrostatin-1 (NCS-1) 68

2.27. Statistical analysis 69

2.27.1. Significance levels 69

2.27.2. Comparing % cell survival, protein expression or viral

titre between sample means 69

2.27.3. Comparing YAP1 protein expression in human tissue

samples 69

2.27.4. Determining an IC50 value by the median-effect

equation 70

2.27.5. The Chou and Talalay equation for measuring the

interaction between two agents at fixed-dose ratios 70

2.27.6. The Bliss Independence equation for measuring the

interaction between two agents at non-fixed dose ratios 71

CHAPTER 3 72

3. TESTING TARGET GENES THAT MAY INFLUENCE THE

SUSCEPTIBILITY OF SCCHN CELL LINES TO REOVIRUS

ONCOLYSIS

73

3.1. INTRODUCTION 73

3.2. STUDY OBJECTIVE 78

3.3. RESULTS 79

3.3.1. Host cell mRNA expression of 8 genes directly correlated

with reovirus IC50 in SCCHN cell lines 79

3.3.2. Validation of SCCHN cell line susceptibility to reovirus-

induced cell death 82

3.3.3. Validation of mRNA expression in the 8 genes in SCCHN

cell lines 85

3.3.4. Optimisation of siRNA-transfection conditions in the PJ41

reovirus-resistant cell line using the KDAlert™ GAPDH

assay kit

87

3.3.5. siRNA-mediated knock-down of the 8 target genes in the

PJ41 cell line 89

3.3.6. siRNA-mediated knock-down of YAP1 sensitised the PJ41

cell line to reovirus-induced cell death 93

3.3.7. siRNA-mediated knock-down of the YAP1 protein in the

PJ41 cell line 100

3.4. DISCUSSION 102

3.5. CONCLUSION 107

IX

CHAPTER 4 108

4. TARGETING YES-ASSOCIATED PROTEIN-1 (YAP1) AS A

FACTOR THAT INFLUENCES REOVIRUS ONCOLYSIS IN

SCCHN CELL LINES

109



4.3. RESULTS 115

4.3.1. Transient over-expression of YAP1 in the PJ41 SCCHN

cell line 115

4.3.2. Transient YAP1 over-expression caused increased

resistance of the PJ34 cell line to reovirus-induced cell

death

118

4.3.3. Stable over-expression of YAP1 in the HN5 cell line 121

4.3.4. Stable over-expression of YAP1 caused increased

resistance to reovirus-mediated cell death in the HN5 cell

line

125

4.3.5. Transient over-expression of YAP1 in the non-cancerous

COS-1 monkey fibroblast cell line 130

4.3.6. Transient YAP1 over-expression caused increased

resistance of the non-cancerous COS-1 cell line to

reovirus-induced cell death

132

4.3.7. Cellular localisation of YAP1 in PJ34 and PJ41 SCCHN

cell lines 135

4.3.8. Treatment of the PJ41 cell line with Sphingosine-1-

phosphate (S1P) caused sensitisation to reovirus oncolysis 138

4.4. DISCUSSION 142

4.5. CONCLUSION 148

CHAPTER 5 149

5. MECHANISTIC STUDIES BEHIND THE INFLUENCE OF YAP1

ON REOVIRUS ONCOLYSIS IN SCCHN CELL LINES 150



5.3. RESULTS 156

5.3.1. Total YAP1 protein expression fluctuates after infection

with reovirus in SCCHN cell lines and in stable over-

expressing YAP1 cell lines

156

5.3.2. JAM-A protein expression did not correlate with the

susceptibility of SCCHN cell lines to reovirus oncolysis,

and was not altered by stable over-expression of YAP1

158

5.3.3. Reovirus protein can be detected in the cytoplasm of

resistant and sensitive SCCHN cell lines 160

X

5.3.4. The rate of intracellular reovirus protein production in

PJ34, HN5 and PJ41 SCCHN cell lines did not correlate

with their reovirus IC50 values

163

5.3.5. Intracellular reovirus protein production was hindered by

stable over-expression of YAP1 167

5.3.6. Extracellular reovirus secretion was indistinguishable in

the SCCHN cell lines and was not hindered by stable

over-expression of YAP1

173

5.3.7. IFN-β secretion from SCCHN cell lines correlated with

their respective reovirus IC50 values, but was not

consistently altered by over-expression or knock-down of

YAP1

175

5.3.8. Detection of the YAP1 protein in SCCHN tissue and

normal tissue 180

5.4. DISCUSSION 190

5.5. CONCLUSION 198

CHAPTER 6 199

6. COMBINING REOVIRUS WITH CHEMOTHERAPEUTIC

TAXANE DRUGS IN PCa CELL LINES 200



6.3. RESULTS 203

6.3.1. Determination of reovirus, Cabazitaxel and Docetaxel

IC50 values in PCa cell lines 203

6.3.2. Concurrent combination of reovirus with Docetaxel or

Cabazitaxel mostly had an anti-cancer synergistic effect

in PCa cell lines, as determined by the Chou and Talalay

equation

207

6.3.3. Concurrent combination of reovirus and Cabazitaxel

resulted in greater synergistic anti-cancer activity in the

DU145 PCa cell line than sequential combination

treatment

212

6.3.4. Combination of reovirus and Cabazitaxel or Docetaxel at

doses below the IC50 values had an anti-cancer

synergistic effect on the DU145 PCa cell line, as

determined by the Bliss Independence model

215

6.3.5. The synergistic anti-cancer effect of the combination of

reovirus and Cabazitaxel or Docetaxel at the IC50 doses

may be due to enhanced microtubule stability

221

6.3.6. The synergistic anti-cancer effect of reovirus in

combination with low doses of Cabazitaxel or Docetaxel 223

XI

may be due to increased microtubule stabilisation

6.3.7. The synergistic anti-cancer effect of reovirus and

Cabazitaxel or Docetaxel in combination was not due to

enhanced viral replication

226

6.3.8. Apoptosis contributes to synergistic cell death when high

doses of the taxane drugs are used in combination with

reovirus, but not at low doses

229

6.4. DISCUSSION 231

6.5. CONCLUSION 237

CHAPTER 7 238

7. GENERAL DISCUSSION 239

7.1. Yes-associated protein-1 (YAP1) as a biomarker of reovirus

treatment response in SCCHN 240

7.2. Combination of reovirus with metronomic doses of taxane drugs

in PCa cell lines 242

7.3. Future work 243

7.3.1. Short-term 243

7.3.2. Long-term 245

REFERENCES 247

XII

LIST OF ABBREVIATIONS

ABC Avidin-Biotin Complex

ABCP Apicobasal Cell Polarity

ADT Androgen Deprivation Therapy

AFP Alpha-Fetoprotein

ALK Anaplastic Lymphoma Kinase Gene

AMOT Angiomotin

AMV Avian Myeloblastosis Virus

ANOVA Analysis of Variance

ANT Adenine Nucleotide Translocase

AR Androgen Receptor

AREG Amphiregulin Gene

ATCC American Type Culture Collection

ATM Ataxia Telangiectasia Mutated Gene

ATP Adenosine Triphosphate

ATV Autologous Tumour Cell Vaccine

Bak Bcl-2 Homologous Antagonist/Killer

Bax Bcl-2-associated X Protein

BCA Bicinchoninic Acid

Bcl-2 B-cell Lymphoma 2

Bcl-xL B-cell Lymphoma Extra Large

Β-hCG Beta-human Chorionic Gonadotropin

Bid BH3 Interacting-Domain Death Agonist

Bim Bcl-2-like protein 11

bp Base Pair

BRAF v-Raf Murine Sarcoma Viral Oncogene Homolog B Gene

BRCA1/2 Breast Cancer 1/2 Gene

BSA Bovine Serum Albumin

CA-125 Carcinoma Antigen 125

CAM Chorioallantoic Membrane

CDK2N2A Cyclin Dependent Kinase 2N2A Gene

cDNA Complementary Deoxyribonucleic Acid

CEC Circulating Endothelial Cell

cFLIP Cellular FLICE Inhibitory Protein

CgA Chromogranin A

CHOP C/EBP Homologous Protein Gene

Ci Confidence Interval

CI Combination Index

CMV Cytomegalovirus

CPE Cytopathic Effect

CR Complete Response

CRB Crumbs Complex

CRPC Castration-Resistant Prostate Cancer

XIII

CRUK Cancer Research UK

Ct Cycle Threshold

CTGF Connective Tissue Growth Factor Gene

CVA21 Coxsackie Virus A21

CVB3 Coxsackie Virus B3

CYLD Cylindromatosis

DAB 3,3'-Diaminobenzidine

DAF Decay-Accelerating Factor

DEPC Diethylpyrocarbonate

DISC Death-Inducing Signalling Complex

DLT Dose Limiting Toxicity

DMEM Dulbecco's Modified Eagle's Medium

DMSO Dimethyl Sulphoxide

DNA Deoxyribonucleic Acid

dNTP Deoxynucleotide Triphosphate

DR Death Receptor

dsDNA Double-Stranded Deoxyribonucleic Acid

dsRNA Double Stranded Ribonucleic Acid

DTT Dithiothreitol

EBV Epstein-Barr Virus

EC Endothelial Cell

ECACC European Collection of Authenticated Cell Cultures

E.Coli Escherichia Coli

ED Effective Dose

EDTA Ethylenediaminetetraacetic Acid

EGF Epidermal Growth Factor

EGFR Epidermal Growth Factor Receptor

eIF Eukaryotic Initiation Factor

ELISA Enzyme-Linked Immunosorbent Assay

EMT Epithelial-Mesenchymal Transition

EnAd Enadenotucirev

EPC Endothelial Progenitor Cell

ER Endoplasmic Reticular

ETS E-Twenty-Six

EV Empty Vector

EYFP Enhanced Yellow Fluorescent Protein

FACS Fluorescence-Activated Cell Sorting

F-Actin Filamentous Actin

FADD Fas-Associated Death Domain

FasL Fas Ligand

FBS Fetal Bovine Serum

FDA Food and Drug Administration

FR Folate Receptor

G418 Geneticin Disulfate Salt

XIV

GADD34 Growth Arrest and DNA Damage-Inducible Protein Gene

GAP GTPase-activating Protein

GAPDH Glyceraldehyde-3-Phosphate Dehydrogenase

GDP Guanosine Diphosphate

gDNA Genomic Deoxyribonucleic Acid

GEF Guanine Nucleotide Exchange Factor

GF Growth Factor

GFP Green Fluorescent Protein

GM-CSF Granulocyte Macrophage Colony-Stimulating Factor

GPCR G-Protein Coupled Receptor

GRP78 78 kDa Glucose-Regulated Protein Gene

GTP Guanosine Triphosphate

GTPase GTP-binding Hydrolysing Protein

h Human

HBV Hepatitis B Virus

HE4 Human Epididymis Protein 4

HER2 Human Epidermal Growth Factor Receptor 2 Gene

HMGB-1 High Mobility Group Box-1

H&N Head and Neck

HPV Human Papillomavirus

HRP Horseradish Peroxidase

HSV Herpes Simplex Virus

HTLV-1 Human T Lymphotropic Virus Type 1

IC50 50% Inhibitory Concentration

ICAM-1 Intercellular Adhesion Molecule-1

ICR Institute of Cancer Research

IFITM3 Interferon-Inducible Transmembrane Protein 3

IFN Interferon

Ig Immunoglobulin

IGF-1 Insulin-like Growth Factor-1

IHC Immunohistochemistry

IL Interleukin

ILT2 Immunoglobulin-like Transcript 2

Ing4 Inhibitor of Growth 4

IRF-3/9 Interferon-Regulatory 3/9

ISG Interferon Stimulated Gene

ISRE Interferon-Stimulated Response Element

ISVP Infectious Subvirion Particle

IT Intratumoural

IV Intravenous

JAK1 Janus Kinase-1

JAM-A Junction Adhesion Molecule-A

JNK c-jun N-terminal Kinase

kDa Kilodalton

XV

KRAS V-Ki-ras2 Kirsten Rat Sarcoma Viral Oncogene Homolog

Gene

KS Kaposi Sarcoma

KSHV Kaposi-Sarcoma-Associated Herpesvirus

L Large

LATS1/2 Large Tumour Suppressor-1/2

LB Lysogeny Broth

LPA Lysophosphatidic Acid

M Medium

MAPK Mitogen-Activated Protein Kinase

MC Metronomic Chemotherapy

MDA5 Melanoma Differentiation-Associated Protein-5

MDR1 Multidrug Resistance Protein-1

MED Minimum Effective Dose

MEK Mitogen-Activated Protein Kinase Kinase

MEM Modified Eagle's Medium

MES 2-(N-morpholino)ethanesulfonic acid

MFI Mean Fluorescence Intensity

MHC Major Histocompatibility Complex

mL Millilitre

MOB1A/1B Mps One Binder Kinase Activator 1A and 1B

MOI Multiplicity of Infection

MOPS 3-(N-morpholino)propansulfonic acid

mRNA Messenger RNA

MST1/2 Mammalian STE20-like Protein Kinase-1/2

MTD Maximum Tolerated Dose

mTOR Mechanistic Target of Rapamycin

MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-

(4-sulfophenyl)-2H-tetrazolium, inner salt]

NARA Neutralising Anti-Reovirus Antibodies

NAT Norma Adjacent Tissue

NCS-1 Necrostatin-1

NDV Newcastle Disease Virus

NF2 Neurofibromatosis Type 2 / Merlin Gene

NF-κB Nuclear Factor Kappa B

NK Natural Killer

NSCLC Non-Small Cell Lung Cancer

NSE Neuron-specific Enolase

OBF Optimal Balance Factor

OD Optical Density

OV Oncolytic Virus

p53 Tumour Protein 53

p73 Tumour Protein 73

PAP Prostatic Acid Phosphatase

XVI

PARP Poly ADP-ribose Polymerases

PBMC Peripheral Blood Mononuclear Cell

p53BP-2 Tumor Suppressor p53-binding Protein 2

PBS Phosphate Buffered Saline

PCa Prostate Cancer

PCR Polymerase Chain Reaction

PD Progressive Disease

PDL-1 Programmed Death Ligand 1

PES Phenazine Ethosulfate

pfu Plaque Forming Unit

PI3K Phosphatidylinositol-3 Kinase

PKR Protein Kinase R

PPxY motif Proline-Proline-x–Tyrosine motif

PR Partial Response

pRb Retinoblastoma Protein

PRR Pattern Recognition Receptor

PSA Prostate Specific Antigen

PSMA Prostate Membrane-Specific Antigen

PTEN PI3K-phosphatase and Tensin Homolog Deleted on

Chromosome 10

Puma p53 upregulated modulator of apoptosis

pYAP Phosphorylated YAP

qPCR Quantitative Polymerase Chain Reaction

RAF Rapidly Accelerated Fibrosarcoma

RalGEF Ras-related Guanine Nucleotide Exchange Factor

RB Retinoblastoma Gene

RIP1/3 Receptor Interacting Protein 1/3

RISC Ribonucleic Acid-induced Silencing Complex

RNA Ribonucleic Acid

RNAi Ribonucleic Acid Interference

ROS Reactive Oxygen Species

RT-qPCR Real Time-Quantitative Polymerase Chain Reaction

Runx2 Runt-related transcription factor 2

S Small

S89 Serine residue 89


S311 Serine reissue 311


SAV1 Salvador Homologue 1

SCC Squamous Cell Carcinoma

SCCHN Squamous Cell Carcinoma of the Head and Neck

SCG2 Secretogranin 2

SCID/NOD Severe Combined Immunodeficiency/ Non-obese Diabetic

SCRIB Scribble

XVII

Sd Stable disease

SD Standard Deviation

SDS Sodium Dodecyl Sulfate

SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis

SEM Standard Error Mean

SH3 SRC Homology 3 Domain

Shh Sonic hedgehog

siRNA Small/short Interfering Ribonucleic Acid

SK1 Sphingosine Kinase-1

Smac Second Mitochondrion-Derived Activator of Caspases

SMAD2/4 Mothers Against Decapentaplegic homolog 2/4 Gene

SOX4 SRY-related HMG-box-4 Gene

S1P Sphingosine-1-Phosphate

S1PR1 Sphingosine-1-Phosphate Receptor 1

ssRNA Single Stranded Ribonucleic Acid

STAT1/2 Signal Transducers and Activators of Transcription 1/2

STR Short Tandem Repeat

SV40 Simian Virus 40

T3A Type 3 Abney

TAD Transcriptional Activation Domain

TAZ Transcriptional Co-activator with PDZ-binding Motif

TCID50 50% Tissue Culture Infective Dose

T3D Type 3 Dearing

TEAD family TEA Domain-Containing Transcription Factor Family

TERT Telomerase Reverse Transcriptase

TGF-β Transforming Growth Factor-beta

T2J Type 2 Jones

T1L Type 1 Lang

TLR Toll-like Receptor

TMB Tetramethylbenzidine

TNF-α Tumour Necrosis Factor-alpha

TNFR1/2 Tumour Necrosis Factor Receptor 1/2

TNM Tumour, Node, Metastasis

TP53 Tumour Protein 53 Gene

TRAF TNF Receptor-Associated Factor

TRAIL Tumour Necrosis Factor-Related Apoptosis-Inducing Ligand

TSG Tumour Suppressor Gene

T-VEC Talimogene Laherparepvec

Tyk2 Tyrosine Kinase-2

UK United Kingdom

UPR Unfolded Protein Response

US United States

USA United States of America

VEGF-A Vascular Endothelial Growth Factor-A

XVIII

vGPCR Viral G-Protein-Coupled Receptor

VSV Vesicular Stomatitis Virus

WGA Wheat Germ Agglutinin

WNT Wingless-type MMTV Integration Site family

XBP-1 X-box Binding Protein 1 Gene

YAP1/2 Yes-Associated Protein-1/2

YAP1 Yes-Associated Protein-1 Gene

5-FC 5-Fluorocytosine

5-FU 5-Fluorouridine

3-MA 3-Methyladenine

XIX

LIST OF FIGURES

Figure 1.1. A diagram showing the main hallmarks of cancer. 3

Figure 1.2. A diagram illustrating the different head and neck cancer regions. 3

Figure 1.3. Molecular structure of the reovirus virion. 17

Figure 1.4. The reovirus replication cycle. 18

Figure 1.5. The proposed ‘reovirus-Ras’ model of selective oncolysis. 20

Figure 1.6. A schematic representation of metronomic and conventional

chemotherapy. 32

Figure 2.1. Chemical structures of the MTS tetrazolium compound and its

formazan product. 41

Figure 3.1. RT-qPCR analysis of microarray-identified up-regulated genes in 9

SCCHN cell lines 80

Figure 3.2. Pearson correlation coefficient between the relative mRNA

expression of the 8 genes and the reovirus IC50 dilution value in

SCCHN cell lines

81

Figure 3.3. Validation of the susceptibility to reovirus oncolysis in 3 SCCHN

cell lines and in 2 non-cancerous, untransformed cell types. 83

Figure 3.4. mRNA expression validation of the 8 genes in 3 SCCHN cell lines. 86

Figure 3.5. Optimisation of siRNA-mediated transfection conditions in the

PJ41 cell line by the KDalert™ GAPDH assay. 88

Figure 3.6. mRNA expression of the 8 target genes in the PJ41 cell line after

siRNA-mediated knock-down. 91

Figure 3.7. Evaluation of reovirus-induced cell death after siRNA-mediated

knock-down of the 8 target genes in the PJ41 cell line. 99

Figure 3.8. The transfection-associated toxicity in each treatment condition. 99

Figure 3.9. YAP1 protein detection in the PJ41 cell line after YAP1 siRNA-

mediated knock-down. 101

Figure 4.1. A schematic representation of the proteins involved in the

mammalian Hippo pathway. 111

Figure 4.2. Functional domains of YAP1 and YAP2; the two major isoforms of

the YAP protein. 112

Figure 4.3. YAP1 mRNA expression in the PJ34 SCCHN cell line after 116

XX

transient over-expression of YAP1.

Figure 4.4. YAP1 protein expression in the PJ34 cell line after transient over-

expression of YAP1. 117

Figure 4.5. The transfection-associated toxicity in the PJ34 cell line. 119

Figure 4.6. Evaluation of reovirus-induced cell death after plasmid-mediated

over-expression of YAP1 in the PJ34 cell line. 120

Figure 4.7. YAP1 protein expression in HN5 SCCHN cell line clones after

stable over-expression of YAP1 using the EYFP-YAP1 plasmid. 123

Figure 4.8. YAP1 protein expression in HN5 SCCHN cell line clones after

stable over-expression of YAP1 using the Flag-YAP1 plasmid. 124

Figure 4.9. The differences in proliferation rates between the stable clones and

the HN5 parental cell line. 127

Figure 4.10. Stable over-expression of the YAP1 protein using the EYFP-YAP1

plasmid, promoted resistance to reovirus in the HN5 SCCHN cell

line.

128

Figure 4.11. Stable over-expression of the YAP1 protein using the Flag-YAP1

plasmid, promoted resistance to reovirus in the HN5 SCCHN cell

line.

129

Figure 4.12. YAP1 protein expression in the COS-1 cell line after transient over-

expression of YAP1. 131

Figure 4.13. The transfection-associated toxicity in COS-1 cells. 133

Figure 4.14. Plasmid-mediated over-expression of YAP1 increased the resistance

of the COS-1 cell line to reovirus oncolysis. 134

Figure 4.15. Immunofluorescent staining of PJ41 and PJ34 SCCHN cell lines for

the YAP1 protein. 136

Figure 4.16. Intensity profiles of total YAP1 and phospho-YAP-S127 in the

PJ41 cell line. 137

Figure 4.17. Treatment of the PJ41 SCCHN cell line with S1P caused

sensitisation to reovirus-mediated cell death. 140

Figure 4.18. Treatment of the PJ41 SCCHN cell line with S1P failed to de-

phosphorylate YAP. 141

Figure 5.1. The YAP1 protein fluctuates after infection with reovirus in the

SCCHN and stable cell lines. 157

XXI

Figure 5.2. JAM-A expression does not correlate with the level of reovirus

oncolysis in SCCHN cell lines, and stable over-expression of YAP1

does not alter the level of JAM-A expression at the cell surface.

159

Figure 5.3. Reovirus infected both sensitive (PJ34) and resistant (PJ41)

SCCHN cell lines at different multiplicities of infection (MOI). 161

Figure 5.4. Reovirus can infect HN5 parental cells and the more resistant-

EYFP-YAP1 stable cell line at different multiplicities of infection

(MOI).

162

Figure 5.5. Infectious intracellular reovirus yield in PJ34, HN5 and PJ41

SCCHN cell lines did not correlate with their reovirus IC50 values,

as determined by the 50% tissue culture infective dose (TCID50)

assay.

165

Figure 5.6. Total intracellular reovirus protein production did not correlate with

the susceptibility to reovirus oncolysis in PJ34, HN5 and PJ41

SCCHN cell lines, as determined by western blotting.

166

Figure 5.7. The rate of infectious intracellular reovirus was hindered by stable

over-expression of YAP1 in the HN5 SCCHN cell line, as

determined by the 50% tissue culture infective dose (TCID50) assay.

169

Figure 5.8. The rate of total intracellular reovirus protein production was

hindered by stable over-expression of YAP1 in the HN5 SCCHN

cell line, as determined by western blotting.

170



cell line, as determined by western blotting.

171



cell line, as determined by flow cytometry.

172

Figure 5.11. There was little difference in extracellular reovirus secretion in

PJ34 and PJ41 SCCHN cell supernatants, or in HN5 parental and

stable YAP1 over-expressing cell supernatants.

174

Figure 5.12. The levels of IFN-β secreted by PJ34, HN5 and PJ41 SCCHN cell

lines after infection with reovirus correlated with their sensitivities

to reovirus oncolysis.

177

XXII

Figure 5.13. Stable over-expression of YAP1 in the HN5 SCCHN cell line did

not consistently increase the levels of IFN-β secretion after

infection with reovirus.

178

Figure 5.14. siRNA-mediated knock-down of YAP1 in the PJ41 SCCHN cell

line did not decrease the levels of IFN-β secretion after infection

with reovirus.

179

Figure 5.15. Optimisation of the enzymatic immunohistochemistry (IHC)

staining protocol in prostate cancer tissue for the detection of the

YAP1 protein.

182

Figure 5.16. Examples of YAP1 protein expression from the FDA999c normal

tissue array and the HN803a head and neck cancer tissue array,

using enzymatic immunohistochemistry (IHC).

183

Figure 6.1. Dose-response curves and IC50 values generated from cells infected

with reovirus alone. 204

Figure 6.2. Dose-response curves and IC50 values generated from cells infected

with Cabazitaxel alone. 205

Figure 6.3. Dose-response curves and IC50 values generated from cells treated

with Docetaxel alone. 206

Figure 6.4. Concurrent combination of reovirus and Cabazitaxel at doses 4, 2,

1, 0.5 and 0.25 fold the IC50 had a synergistic anti-cancer effect in

the DU145 PCa cell line.

208

Figure 6.5. Concurrent combination of reovirus and Docetaxel at doses 4, 2, 1,

0.5 and 0.25 fold the IC50 had a synergistic anti-cancer effect in the

DU145 PCa cell line.

209

Figure 6.6. Concurrent combination of reovirus and Cabazitaxel at doses 4, 2,

1, 0.5 and 0.25 fold the IC50 had a synergistic anti-cancer effect in

the LNCaP PCa cell line.

210

Figure 6.7. Concurrent combination of reovirus and Docetaxel at doses 4, 2, 1,

0.5 and 0.25 fold the IC50 had a synergistic anti-cancer effect in the

LNCaP PCa cell line at the ED50 and ED75, but not at ED90.

211

Figure 6.8. Concurrent combination treatment of reovirus and Cabazitaxel

resulted in a more efficient anti-cancer synergistic interaction than

sequential combination treatment, in the DU145 PCa cell line.

214

XXIII

Figure 6.9. Combination treatment of the DU145 PCa cell line with reovirus

and Cabazitaxel at doses much lower than the IC50 values mostly

had an anti-cancer synergistic effect, as determined by Bliss

Independence analysis.

218


and Docetaxel at doses much lower than the IC50 values had an anti-

cancer synergistic effect, as determined by Bliss Independence

analysis.

220

Figure 6.11. Microtubule stability was enhanced after combination treatment

with reovirus and Cabazitaxel or Docetaxel at the IC50 doses, in the

DU145 PCa cell line.

222


with reovirus and Cabazitaxel at low doses, in the DU145 PCa cell

line.

224


with reovirus and Docetaxel at low doses, in the DU145 PCa cell

line.

225


and Cabazitaxel did not enhance the intracellular or extracellular

viral yield compared to single agent reovirus treatment.

227


and Docetaxel did not enhance the intracellular or extracellular viral

yield compared to single agent reovirus treatment.

228

Figure 6.16. High doses of reovirus in combination with Cabazitaxel or

Docetaxel causes synergistic anti-cancer cell death by apoptosis,

but low dose combination treatment is non-apoptotic.

230

Figure 7.1.

Potential mechanism of YAP1-mediated resistance to reovirus

oncolysis in SCCHN cell lines 244

XXIV

LIST OF TABLES

Table 1.1. Clinical and biological characteristics of HPV negative and positive

SCCHN. 4

Table 1.2. A selection of oncolytic viruses in pre-clinical studies. 14

Table 1.3. A selection of oncolytic viruses in clinical studies. 15

Table 1.4. The 11 viral proteins encoded by the 10 dsRNA segments of the

reovirus genome. 16

Table 1.5. Tumour markers of treatment response currently in use in the UK. 23

Table 1.6. A summary of the Reolysin® clinical trials that have been completed

or are currently ongoing. 25

Table 2.1. Reovirus and Taxane drugs 36

Table 2.2. Cell Culture Media used in this study. 36

Table 2.3. Cell lines used in this study, their growth media, tissue type and

source. 37

Table 2.4.

The forward and reverse primer sequences of all target genes used in

the RT-qPCR reaction. 43

Table 2.5. Cell seeding densities used. 44

Table 2.6. Volumes of siPORT NeoFX and Opti-MEM used per well. 44

Table 2.7. Volumes of each component of the KDalert™ master mix used per

well 45

Table 2.8. The siRNAs used in this study. Two different siRNAs were used for

each target gene. 46

Table 2.9. Primary and secondary antibodies used for western blotting. 50

Table 2.10. The preparation of plasmid DNA and the ‘no plasmid DNA’ control. 52

Table 2.11. The preparation of Lipofectamine® LTX. 52

Table 2.12. The DNA plasmids used for transient or stable transfection of cell

lines. 54

Table 2.13. Primary and secondary antibodies used for immunofluorescence

staining. 56

Table 2.14. Primary and secondary antibodies used for indirect flow cytometry

protein detection. 59

XXV

Table 2.15. Recommended symbols for describing the interaction between

reovirus and Cabazitaxel or Docetaxel when analysed by the CI

equation of Chou and Talalay.

71

Table 3.1. Professor Kevin Harrington’s laboratory showed that 9 SCCHN cell

lines had different sensitivities to reovirus-induced cell death. 74

Table 3.2. Summary of the 8 genes identified as potential predictors of

susceptibility to reovirus oncolysis in SCCHN cell lines and their

corresponding protein functions.

76

Table 3.3. Reovirus IC50 values of 3 SCCHN cell lines, the MRC-5 human

fibroblast cell line and PBMCs isolated from a healthy donor, were

calculated using CalcuSyn software.

84

Table 3.4. A summary of the % knock-down of each target gene in the PJ41 cell

line by transfection with 2 different siRNAs. 92

Table 5.1. Demographic data from the HN803a head and neck cancer tissue

array, with details of IHC YAP1 intensity scoring and cellular

localisation.

184

Table 5.2. Demographic data from the FDA999c multiple organ normal tissue

array, with details of IHC YAP1 intensity scoring and cellular

localisation.

186

Table 5.3. Statistical comparisons of YAP1 protein expression in the tissue

sections by the Chi-squared (χ2) statistical test. 188

Table 5.4. Statistical comparison of immunohistochemistry (IHC) YAP1

staining intensity in the head and neck (H&N) carcinoma tissue

sections by the Chi-squared (χ2) statistical test.

189

1

CHAPTER 1

INTRODUCTION

2

1. INTRODUCTION

1.1. CANCER

1.1.1. Hallmarks

According to Hanahan and Weinberg, there are six main hallmarks that govern the

transformation of a normal cell to a neoplastic state. There are also emerging

hallmarks such as the ability of cancer cells to evade immune destruction (Figure 1.1)

[1]. Cancer cells can sustain chronic proliferative signalling by synthesising growth

factor (GF) ligands, to which they can become hyper-responsive to by over-

expression or structural alteration of associated cell-surface receptors [1]. In normal

cells, tumour suppressor genes (TSGs) function to halt cell cycle progression if

exposed to genomic damage, and if un-repairable, induce apoptosis. Two TSGs that

are commonly inactivated in human cancers are TP53 and RB, which encode tumour

protein 53 (p53) and retinoblastoma (pRb) proteins respectively [1].

Tumour cells have developed a number of strategies to evade programmed cell death,

for example, through increased expression of the anti-apoptotic factors Bcl-2 or Bcl-

xL, or by down-regulation of the pro-apoptotic factors Puma, Bax or Bim [1]. During

autophagy, cytoplasmic constituents are degraded and recycled, and in doing so,

release metabolites that provide energy to the stressed cell [1]. Hence, autophagy can

paradoxically mediate tumour cell survival and death [2]. In contrast, necrotic cells

rupture to spill their contents into the tissue microenvironment. Recruitment of

inflammatory immune cells to remove necrotic debris may benefit surviving tumour

cells by providing growth-stimulatory factors [1].

Normal cells have a limited number of growth and division cycles before they die [1].

However, cells can occasionally overcome the state of crisis and become

immortalised, for example, through over-expression of the DNA polymerase enzyme,

telomerase [1]. In healthy cells, angiogenesis is transiently activated in response to

physiological processes such as wound healing. Tumours may obtain nutrients and

oxygen to sustain growth by inducing angiogenesis permanently, often via up-

regulation of vascular endothelial growth factor-A (VEGF-A) [1]. Another hallmark

of cancer is the activation of invasion and metastasis, which is frequently stimulated

by a process called epithelial-mesenchymal transition (EMT) [1].

3

Figure 1.1. A diagram showing the main hallmarks of cancer. Adapted from [1].

1.1.2. Head and Neck Cancer

1.1.2.1. Incidence

Human head and neck cancers can occur in the oral cavity, pharynx, larynx, salivary

gland, paranasal sinuses and nasal cavity, as depicted in Figure 1.2 [3]. In the United

Kingdom (UK) in 2013, there were 7,591 new cases, making it the 14th most common

cancer [4]. Worldwide in 2012, lip and oral cavity cancer was the 15th leading cancer

by incidence, accounting for more than 300,000 newly diagnosed cases [4]. These

cancers usually manifest in the fifth, sixth or seventh decades of life, and are twice as

common in men as in women [4].

Figure 1.2. A diagram illustrating the different head and neck cancer regions [3].

http://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&uact=8&docid=XhbQsBmIH1eJeM&tbnid=kZbElbgFhRBpFM:&ved=0CAUQjRw&url=http://pubweb.fccc.edu/cancerconversations/2013/04/neck-dissection-not-so-radical-any-more/&ei=8dzIU_HnI6ef0QXmo4B4&bvm=bv.71198958,d.d2k&psig=AFQjCNEBPP3Hp1w5vcNINKKS-fgtlMadIQ&ust=1405759085937612

4

Approximately 90% of these cancers are biologically similar and originate from the

squamous cells that line the mucosal surfaces. These cancers are therefore called

squamous cell carcinomas of the head and neck (SCCHN) [3].

1.1.2.2. Risk factors and symptoms

The two most important risk factors are tobacco and alcohol use. Infection with

human papillomavirus (HPV) is also a major risk factor, especially for oropharyngeal

cancers [3-5]. Other risk factors may include genetic predisposal via inherited

disorders such as Fanconi anaemia, radiation exposure, and occupational exposure to

wood dust, asbestos or synthetic fibres [3-5]. General symptoms may include a

persistent sore throat, a sore that does not heal, hoarseness in the voice and difficulty

in swallowing. There may be a persistent earache or frequent headaches [3, 4].

1.1.2.3. Heterogeneity

SCCHN is a heterogeneous disease. The HPV-status of a SCCHN tumour is

considered to be important in predicting a patient’s prognosis [5]. HPV is a double-

stranded DNA (dsDNA) virus that contains two oncogenes, E6 and E7, whose

expression inactivates the p53 and pRb tumour suppressors respectively [5]. This is

believed to be the onset of HPV-mediated carcinogenesis in some SCCHNs.

Interestingly, HPV-positive SCCHN cancers are typically TP53 wild type and occur

predominantly in patients without a history of tobacco and/or alcohol use. In contrast,

TP53 mutations are frequently observed in smoking and/or alcohol-related HPV-

negative SCCHN tumours, and these patients have a less favourable prognosis [5].

These characteristics are summarised in Table 1.1.

Table 1.1. Clinical and biological characteristics of HPV negative and positive SCCHN. Adapted

from [5].

Feature HPV-negative SCCHN HPV-positive SCCHN References

Incidence Decreasing Increasing [6]

Aetiology Smoking, excessive alcohol use Oral sex [7]

Age Above 60 years Under 60 years [6]

Field cancerisation Yes Unknown [8]

TP53 mutations Frequent Infrequent [9]

Predilection site None Oropharynx [10]

Prognosis Poor Favourable [11, 12]

5

In addition to abrogation of the p53 and pRb signalling pathways, cell cycle

regulation can be perturbed by inactivation of CDKN2A (encoding the p16INK4A

tumour suppressor protein) [13] or over-expression of CCND1 (encoding Cyclin D1; a

protein that regulates cell cycle progression) [9], in many SCCHN tumours. Increased

activity of telomerase or its catalytic subunit, telomerase reverse transcriptase

(TERT), is frequently detected [5]. A subgroup of SCCHNs over-express the

epidermal growth factor receptor gene (EGFR) giving rise to enhanced proliferation

[5, 14], whereas other SCCHNs escape from the tumour suppressing effects of the

transforming growth factor-β (TGF-β) pathway by somatic mutation or chromosomal

loss of certain genes, including SMAD2 and SMAD4 [15]. Also linked to the

development of 10-20% of SCCHNs are activating mutations in the

phosphatidylinositol-3 kinase (PI3K) -PI3K-phosphatase and tensin homolog deleted

on chromosome 10 (PTEN) -Akt pathway. For example, somatic mutations may

occur in PIK3CA and PTEN genes that effectively reduce apoptosis and allow

proliferation [5, 16, 17].

1.1.2.4. Pathogenesis

Evidence suggests that SCCHN may arise from a precursor lesion known as a

leukoplakia. Surgical removal or treatment of the leukoplakia is largely ineffective in

preventing malignant transformation and hence, clinicians normally employ watchful

waiting [5]. Additionally, various genetic precursor changes arise in the squamous

epithelium that are not visible to the naked eye. Such changes are referred to as field

cancerisation, and may include a mutated p53, decreased cytokeratin 4 and loss of

heterozygosity at chromosome 9p. It is thought that this may be the source of local

recurrences and second primary tumours after resection of the original SCCHN [5].

1.1.2.5. Diagnosis, treatment and prognosis

The most common method for staging head and neck cancers is the tumour, node,

metastasis (TNM) system [18, 19]. ‘T’ describes the characteristics of the tumour at

the primary site, which may be based on size, location or both. ‘N’ indicates the

degree of regional lymph node involvement, and ‘M’ describes the absence or

presence of distant metastases. The specific TNM status of each patient is then

tabulated to give a numerical status of Stage I, II, III, or IV. In general, early-stage

6

disease is denoted as Stage I or II and advanced-stage disease as Stage III or IV.

SCCHN patients that present with early-stage disease can be treated with surgery or

radiotherapy and generally have a favorable prognosis [5]. However, up to 50% of

SCCHN patients present with advanced disease [20]. The standard treatment for

advanced cases is surgery combined with chemotherapy (typically Paclitaxel,

Carboplatin or Docetaxel) and radiotherapy, and more recently the use of the EGFR-

inhibitor, Cetuximab. Unfortunately, the relapse rate at 2 years for patients with

locally advanced SCCHN is 30-50%, as loco-regional recurrences, distant metastasis

and second primary tumours often develop [5, 21]. The patients in the relapsed and

metastatic group have an overall 5-year survival rate of less than 10% [21].

Therefore, new therapeutic strategies are needed. Oncolytic viruses are biological

anti-cancer agents that may be an attractive option for SCCHN, as discussed in

Section 1.2 and 1.3.

1.1.3. Prostate Cancer

1.1.3.1. Incidence

In 2011 there were 41,736 new cases of prostate cancer (PCa) in the UK, making it

the most common cancer in men. Worldwide, it is the fourth most common cancer

overall, with more than 1,111,000 new cases diagnosed in 2012 [4]. The incidence of

PCa rises from the ages of 50-54, reaching a peak in the 75-79 age group [4]. Acinar

adenocarcinoma is the most common type, amounting to more than 90% of cases [22].

1.1.3.2. Risk factors and symptoms

Deleterious mutations in the BRCA1 and BRCA2 TSGs have been shown to be

important risk factors in PCa development [23, 24]. Men with Lynch syndrome, an

autosomal dominant genetic condition, are also at higher risk than the general

population [22]. Men whose father or brother has/had PCa [25], or whose mother

has/had breast cancer [26], are at greater risk of developing the disease than men

without such a family history. A study published by Hoffman et al concluded that

PCa more commonly affects African American and Hispanic men than non-Hispanic

white men [27]. Elevated circulating levels of insulin-like growth factor-1 (IGF-1)

7

[28, 29] and testosterone [22, 30] have been linked to an increased risk of PCa. A

small number of studies associate occupational exposure to arsenic, ionising radiation

or rubber, as possible risk factors [22, 30]. Early PCa generally does not cause any

symptoms. However, some PCa growths may cause symptoms similar to benign

prostatic hyperplasia, such as frequent urination, haematuria and dysuria [22].

1.1.3.3. Heterogeneity and pathogenesis

The development and progression of PCa is usually dependent on androgen receptor

(AR) signalling. Testosterone is the main circulating androgen hormone produced by

the testes and is converted by 5α-reductase into dihydrotestosterone. The binding of

androgens to their cognate AR gives rise to a conformational change that causes the

receptor to dissociate from heat shock proteins in the cytoplasm [31, 32]. The

receptor then translocates to the nucleus, where it acts as a transcription factor and

binds to androgen response elements in the promoter regions of target genes, such as

prostate specific antigen (PSA). Transcription and expression of these target genes

are essential for both normal prostate development and prostate carcinogenesis [31,

32]. Therefore, PCa patients are usually treated with androgen deprivation therapy

(ADT). However, patients will eventually no longer respond to ADT, relapse, and

develop castration-resistant PCa (CRPC). This may be due to enhanced sensitivity of

the AR to its agonists, AR mutations that cause the receptor to be responsive to other

non-androgen ligands, or ligand-independent AR activation [32]. Like SCCHN, PCa

displays a high level of heterogeneity. There is evidence to suggest that the

transcriptional activity of the AR in CRPC is maintained by up-regulation of growth

factors such as epidermal growth factor (EGF) (and its EGFR receptor) [33], and IGF-

1 [34], as well as the cytokines interleukin-6 and -8 (IL-6 and IL-8) [35-37]. PCa

proliferation and survival may also be mediated by activation of the mitogen-activated

protein kinase (MAPK) and PI3K-PTEN-Akt signalling pathways [38, 39].

Expression of the PTEN tumour suppressor protein was found to be lost in 20-27% of

primary tumours and 79% of CRPC cases [38]. Approximately 50% of prostate

tumours contain gene fusions between the E-twenty-six (ETS) family of transcription

factors and the androgen receptor (AR) gene promoter, which contributes to

neoplastic development [32].

8

1.1.3.4. Diagnosis, treatment and prognosis

A high serum PSA may help indicate whether a man has PCa. According to the

American Cancer Society, a PSA level above 4ng/mL is deemed abnormal, and these

patients are advised to undergo a biopsy [32]. Histopathological analysis of biopsy

tissue (Gleason score) is key in determining a patient’s treatment and prognosis [31],

and the stage of PCa is usually determined by the TNM system [22]. Low risk

localised PCa is unlikely to develop for many years and is monitored by active

surveillance. If the cancer starts to grow then the patient may be offered surgery to

remove the prostate gland or internal radiotherapy to the prostate (brachytherapy)

[22]. For locally advanced cases where the cancer has broken through the capsule

surrounding the prostate gland, the usual treatments are surgery to remove the prostate

gland or external radiotherapy to the prostate [22]. ADT is also given before, during

or after treatment, which usually improves symptoms transiently before tumours

become castration resistant, enabling disease progression. Cryotherapy and high

frequency ultrasound therapy may also be offered, but these are not standard

treatments [22]. The first line of standard care for patients with CRPC is the

chemotherapeutic Docetaxel, which has shown to have a median survival benefit of

up to 3 months [40]. Cabazitaxel, a newer taxane analogue, is used for treatment of

patients with CRPC previously treated with a Docetaxel-containing regimen [22]. For

disease that is confined to the prostate, the 5-year survival for patients in England in

1999-2002 was ≥90%. However, if the disease was metastatic at diagnosis the 5-year

survival dropped to 30% [4, 22], and clearly more efficacious treatments are required.

Oncolytic viruses are one such option (Section 1.2 and 1.3).

9

1.2. ONCOLYTIC VIROTHERAPY

Oncolytic Viruses (OVs) display anti-cancer activity in a wide range of tumour types,

including SCCHN and PCa. They preferentially infect and lyse cancer cells either

naturally, or through modification to include a therapeutic gene that aids targeting of

the tumour. Many different OVs have been investigated in pre-clinical and clinical

settings (as summarised in Table 1.2 and 1.3 respectively) and overall, have a good

safety profile. As a monotherapy, OVs are probably not potent enough to achieve

complete tumour regressions or to generate sustained clinical responses [41].

However, the combination of OVs with conventional cancer treatment modalities

have shown to enhance their efficacy [42]. Naturally occurring OVs are considered to

be safer than their genetically modified counterparts. Moreover, modifications to

reduce the pathogenicity of an OV can also compromise its anti-cancer efficacy.

1.2.1. History

Using viruses for the treatment of cancer is not a new concept. In the early twentieth

century, it was noticed that cancer patients who contracted naturally occurring viral

infections sometimes went into brief periods of clinical remission [43]. Advances in

cell and virus culture techniques in the 1950’s and 1960’s sparked investigation of

different viruses for the treatment of cancer in animal models and in humans [44].

However, poor efficacy led to a temporary halt in oncolytic virotherapy research [45].

In 1991, there was a resurgence of interest following a publication that described the

use of a thymidine kinase-negative mutant of HSV-1 as a possible treatment for

glioma [46]. Since the arrival of recombinant technology, there has been much focus

on engineering viruses to eliminate their pathogenicity without destroying their

oncolytic potency [45]. Six OVs in current use are reviewed below, before focussing

in depth on oncolytic reovirus (Section 1.3).

1.2.2. Oncolytic Adenovirus

Adenoviruses contain dsDNA and are non-enveloped. They have been engineered to

enhance their anti-tumour potency, delivery and safety. Various oncolytic

adenoviruses have been tested in pre-clinical models [21]. A significant anti-tumour

effect was observed in a SCCHN murine tumour model after subcutaneous injection

of Ad5/3-Δ24FCU1 [47]. Ad5/3-Δ24FCU1 features a 5/3 serotype chimeric capsid to

10

enhance gene delivery; a 24-base pair (bp) deletion in the pRb-binding domain of the

viral E1A protein to enable selective replication in tumour cells; and a FCU1 suicide

gene that encodes a protein that catalyses the conversion of 5-fluorocytosine (5-FC) to

the active drug 5-fluorouridine (5-FU), in order to enhance selective cancer cell kill

[21, 47]. Ad5/35 is another genetically modified adenovirus that substitutes Ad5 for

the Ad35 fiber, which recognises the CD46 receptor on the surface of many tumour

cells and thus, increases tumour selectivity. In combination with cisplatin or

radiation, Ad5/35 demonstrated delayed tumour progression in xenograft mice models

of head and neck cancer or melanoma [21, 48, 49]. Arming adenovirus with anti-

angiogenesis genes that encode human endostatin (Ad-Endo) has shown to promote

anti-tumour activity in a nasopharyngeal carcinoma model [21, 50]. OBP-301, an

adenovirus that displays anti-tumour selectivity in cancer cells expressing telomerase,

was shown to overcome radio-resistance in oral squamous cell carcinomas in vivo [21,

51], and supressed LNCaP tumour growth in a PCa mouse model [52]. Hu et al tested

two oncolytic adenoviruses that targeted TGF-β (a regulator of PCa metastasis) in a

PCa mouse model. Significant inhibition of tumour growth and bone metastasis was

observed, indicating the potential of the viruses to treat metastatic PCa [53].

Enadenotucirev (EnAd; previously known as ColoAd1) is an Ad11p/Ad3 chimeric

OV that has specific activity in human colon cancer, and has entered Phase I clinical

testing to examine its therapeutic potential [54]. ONYX-015 is an E1B gene deleted

adenovirus that selectively replicates in and lyses tumour cells with a deficient TP53

[21]. Normal cells with a functioning TP53 gene should remain un-harmed after virus

infection, as replication should be prevented by p53-mediated cell cycle arrest [21,

55]. In patients with malignant glioma, ONYX-015 was injected into the

peritumoural region after surgical removal of the tumour. Results showed that

ONYX-015 was safe but had little therapeutic effect [56]. Conversely, in a Phase II

clinical study, significant tumour regression was observed following peritumoural and

intratumoural injection of ONYX-015 in patients with recurrent head and neck

cancers [21, 57]. However, it was later demonstrated that ONYX-015 could kill

tumour cells regardless of p53 status and therefore, the exact mechanism of its

selectively remains to be found [21]. In 2005, Chinese regulators approved the

modified oncolytic adenovirus H101 for the treatment of head and neck cancer. H101

11

is an E1B-deleted adenovirus that is similar to ONYX-015 but lacks all E3 proteins

[21], and has shown to improve overall response rates in these patients [21].

1.2.3. Oncolytic Herpes Simplex virus

Herpes simplex virus (HSV) is an enveloped dsDNA virus. Infected host cells would

normally initiate an anti-viral response pathway to prevent production of HSV virions

via phosphorylation of protein kinase R (PKR). However, expression of the viral

protein ICP34.5 inhibits the action of PKR by de-phosphorylating its downstream

component, eukaryotic initiation factor-2α (eIF-2α), allowing viral replication to

proceed [21]. Expression of another viral protein, ICP47, allows HSV to evade the

effects of the immune system by inhibiting major histocompatibility complex (MHC)

class I, therefore preventing virus clearance [21]. Deletion of these genes responsible

for viral pathogenicity and immunogenicity has improved the cancer cell selectivity of

HSV and has provided increased safety to normal bystander cells [21].

Talimogene laherparepvec (T-VEC) (originally named Onco-VEXGM-CSF), is a

genetically engineered HSV-1 that secretes granulocyte macrophage colony-

stimulating factor (GM-CSF) to enhance systemic anti-tumour immune responses

[58], and contains deletions in ICP47 and ICP34.5. T-VEC was the first OV in the

western world to complete a Phase III clinical trial, and was approved by the United

States (US) Food and Drug Administration (FDA) in 2015 for the treatment of

melanoma in patients with in-operable tumours. T-VEC was well-tolerated and some

patients showed a complete response, demonstrating its capability to induce a

systemic immune effect that kills distant, un-injected tumours [59]. In a Phase I/II

clinical study, TVEC showed signs of efficacy in patients with advanced stage

SCCHN who were also being treated with cisplatin and radiotherapy [21, 60]. A

naturally occurring mutant of HSV-1 named HF10, lacks the expression of several

functional genes, and caused considerable cancer cell death after intratumoral

injection in recurrent SCCHN tumours [21, 61].

Multiple pre-clinical studies with different HSV mutants have displayed oncolytic

potential, as summarised in Table 1.3. For example, a study involving combination

therapy of two HSV-1 mutants demonstrated enhanced oncolytic activity compared to

using either virus alone concurrently, in an oral squamous cell carcinoma xenograft

mouse model [21, 62]. Additionally, a number of HSV-1 strains have shown promise

12

in treating PCa in in vivo models, including viruses armed with prostatic acid

phosphatase (PAP), interleukin-12 (IL-12) or inhibitor of growth 4 (Ing4) [63].

1.2.4. Oncolytic Vaccinia virus

Vaccinia virus is a large, enveloped member of the poxvirus family, and contains a

linear dsDNA genome [64]. Deletion of non-essential genes increases the selective

replication of oncolytic vaccinia virus in tumour cells and improves safety by

attenuating infection in normal cells. For example, tumours with Ras or p53

mutations are more susceptible to viral replication when the viral gene encoding

thymidine kinase is deleted. GLV-1h68 is an attenuated, replication competent

vaccinia virus that has anti-tumour activity in head and neck cancer cell lines, as well

as in in vivo mouse models of SCCHN [21, 65]. The oncolytic effect of this virus has

also been shown in pancreatic and prostate tumour xenografts [66, 67].

In a Phase I clinical study, a recombinant vaccinia virus expressing PSA (rV-PSA)

provoked PSA-specific immune responses and clinical activity in men with advanced

PCa [68]. An oncolytic thymidine kinase-deleted vaccinia virus called JX-594

expresses gene encoding GM-CSF. This Wyeth strain vector has demonstrated

enhanced anti-tumour immunity and selective oncolytic activity in solid tumours [21].

A Phase III clinical trial aiming to assess the efficacy of JX-594 in combination with

Sorafenib, is now recruiting patients with advanced hepatocellular carcinoma [69].

1.2.5. Oncolytic Newcastle Disease virus

Newcastle Disease virus (NDV) is an enveloped, negative-sense, single stranded (ss)

RNA paramyxovirus that causes fatal disease in birds, but is non-pathogenic in

humans [21]. Oncolytic NDV, for example, the MTH-68/H attenuated strain, has

anti-tumour activity in human cancer cells with aberrant antiviral or apoptotic

signalling pathways [44, 70]. In contrast, normal cells are able to halt viral replication

because they have intact interferon signalling. A vaccine named ATV-NDV

(autologous tumour cell vaccine, modified with non-lytic NDV) has been trialled in

various clinical studies and has so far shown positive results in patients with

colorectal cancer [71], breast cancer, glioblastoma [72] and SCCHN [73]. In order to

enhance the cancer therapeutic efficacy, Vigil et al used reverse genetics to create a

modified NDV containing a fusogenic F protein (NDV/F3aa) that is capable of

forming syncytia. NDV/F3aa caused significant reduction in tumour development in

13

a colon carcinoma xenograft mouse model, compared to mice treated with unmodified

virus [74]. It also exerted potent oncolytic activity against SCCHN cell lines and

murine flank tumours [21, 75]. Interestingly, the combination of Reovirus Type 3

Dearing (T3D) and the Hitcher B1 strain of NDV showed synergistic oncolytic

activity against human glioblastoma cell lines and glioma xenografts [76].

1.2.6. Oncolytic Vesicular Stomatitis virus

Vesicular Stomatitis virus (VSV) is a member of the rhabdoviridae family and has a

negative-sense RNA genome. VSV has shown to have some oncolytic activity, and it

is believed that tumours with defective antiviral responses are most susceptible [63].

There have been concerns over the ability of wild type strains of VSV to replicate in

the central nervous system, and thus, recombinant attenuated VSVs have been

developed to improve oncolysis and safety [63]. VSV has been engineered to express

antibodies for prostate membrane-specific antigen (PSMA), EGFR and folate receptor

(FR), which were shown to replicate specifically in PCa cells expressing the

corresponding target receptor on their cell surface [63, 77]. VSV has also been

genetically engineered to express interferon-β (IFN-β) in order to enhance its efficacy,

and is now being tested in a Phase I liver cancer trial [78].

1.2.7. Oncolytic Coxsackie virus

Coxsackie virus is a non-enveloped, positive-sense, ssRNA virus that belongs to the

picornaviridae family [21]. The oncolytic therapeutic potential of non-engineered

strains such as coxsackie virus B3 (CVB3) has been demonstrated in pre-clinical

cancer models. When administered into the local tumour microenvironment, CVB3

stimulated immunogenic cytotoxicity through the release of adenosine triphosphate

(ATP), calreticulin, and high mobility group box-1 (HMGB-1) [79]. This aided

priming of adaptive immunity through recruitment of dendritic cells and natural killer

(NK) cells [70]. Cancer cells over-expressing the intercellular adhesion molecule-1

(ICAM-1) and decay-accelerating factor (DAF) receptors are susceptible to infection

and lysis of another coxsackie virus, Coxsackie virus A21 (CVA21; CavatakTM) [80].

Anti-tumour activity of CVA21 has been confirmed in various cancer types in the pre-

clinical setting, including multiple myeloma [81], prostate cancer [82], breast cancer

[83], and advanced stage melanoma [84]. A Phase II clinical trial has shown signs of

efficacy after administration of CVA21 in patients with melanoma [85, 86].

14

Table 1.2. A selection of oncolytic viruses in pre-clinical studies. Adapted from [21, 70].

PRE-CLINICAL STUDIES

Virus Strain Mechanism of selectivity Route Tumour type Model Reference

Adenovirus

Ad5/3-Δ24FCU1

The knob fiber of serotype 5 is replaced with the knob of serotype 3, resulting in a 5/3 chimera, which enhances gene delivery and cancer cell killing. Has a 24-bp deletion (Δ24) in

the constant region 2 domain of the viral E1A gene, therefore EIA is unable to bind to pRb for

effective replication in normal cells, and only replicates in cancer cells with defective pRb.

Expresses the FCU1 fusion protein that catalyses the conversion of 5-FC to 5-FU, thus

bypassing resistance of cancer cells to 5-FU.

IT SCCHN Xenograft mice [47]

Ad5/35

Substitution of Ad5 for the Ad35 fiber knob to recognise the CD46 receptor, which is abundant on cancer cells, allowing improved viral entry and anti-tumour activity. E1A

expression is controlled by the tumour specific E2F-1 promoter to limit viral replication to

cancer cells with defective pRb.

IT SCCHN, melanoma Xenograft mice [48, 49]

Ad-Endo Replication deficient due to deletion of viral E1 region and part of the E3 region, for

improved safety. Encodes human endostatin to inhibit tumour angiogenesis. IT SCCHN Xenograft mice [50]

OBP-301

E1A and E1B genes are driven by the human telomerase reverse transcriptase (hTERT)

promoter, which positively regulates telomerase. This replication competent virus is selective for cancer cells with activated telomerase.

IT + radiation

IT

Radio-resistant SCCHN,

PCa

Orthotopic mice

Xenograft mice

[51, 52]

Ad.sTβRFc and TAd.sTβRFc

E1 deleted. Ad.sTβRFc expresses TGF-β receptor II and is fused with human Fc. It reduces

tumour growth by binding to TGF-β, thus inhibiting TGF-β signalling. TAd.sTβRFc

replication is driven by hTERT promoter in cancers with activated telomerase.

Tail vein PCa Orthotopic mice [53]

HSV

HSV-IR849 + HSV-1 HF γ34.5 gene deficiency restricts the ability of the virus to replicate in the adult nervous system. IT SCCHN Xenograft mice [62]

bPΔ6-hPAP

Deleted ICP6 gene restricts virus replication in normal cells, making it selective for cancer

cells. Expresses human prostatic acid phosphatase (PAP) to generate an anti-tumour response.

IT PCa Xenograft mice [87]

NV1042 oHSV

A HSV-1/HSV-2 intertypic recombinant that expresses IL-12 to generate anti-tumour and

anti-angiogenic responses. Retains one copy of the γ34.5 gene, and gene deletion of α47

enhances immunosurveillance of host cell.

IT + Vinblastine PCa Xenograft mice [88]

HSV1716Ing4 HSV1716 contains a gene deletion in ICP34.5, thus improving tumour-selective replication.

Expresses inhibitor of growth 4 (Ing4) to enhance oncolytic potency. IT PCa Xenograft mice [89]

Vaccinia GLV-1h68

Genetically modified by creating interruptions in the thymidine kinase, F14.5L, and

heamagglutinin genes. This decreases virulence in normal cells and enables tumour

selectivity of the virus.

IT

IT

IT

SCCHN

Pancreatic

PCa

Orthotopic mice

Xenograft mice

Xenograft mice

[65-67]

NDV NDV (F3aa)

Conatins the attenuated NDV Hitchner B1 strain (NDV-B1), with the cleavage site of the F

protein modified with three amino acid changes, making it more fusogenic and has the ability

to form syncytia.

IT

IV

SCCHN

Colorectal

Xenograft mice

Xenograft mice

[74, 75]

CVA CVB3, CVA21 Non-engineered strains are selective for cancer via associated cell surface receptors ICAM-1

and DAF. IV

Melanoma, PCa, multiple myeloma,

breast, lung

Xenograft or

orphotopic mice [79, 81-84]

15

Table 1.3. A selection of oncolytic viruses in clinical studies. Adapted from [21, 70].

CLINICAL STUDIES

Virus Strain Mechanism of selectivity Route Tumour type Type of study Reference

Adenovirus

ONYX-015

E1B-55 gene deletion. Selective

replication and lysis of p53-deficient

cancer cells.

IT Various malignancies

including head and neck Phase II [56, 57]

H101 E1B-55 and E3 gene deletion.

IT

IT + chemotherapy

IT + cisplatin and 5-FU

Various malignancies

Various malignancies

Head and neck, oesophageal

Phase I

Phase II

Phase III

[90-92]

HSV T-VEC

ICP34.5 and ICP47 gene deletion and

GM-CSF expression.

IT

IT + chemo-radiation

Breast, gastrointestinal, head

and neck and malignant

melanoma

SCCHN

Phase I/III

Phase I/II

[60, 93]

[59]

HF10 Lack of UL56 protein IT SCCHN, breast, melanoma Phase I [61, 94]

Vaccinia

rV-PSA

Recombinant vaccinia virus expressing

PSA to induce PSA-specific immune

responses.

IT PCa Phase I [68]

JX-594 Thymidine kinase-deleted virus

expressing GM-CSF.

IV

IT

IT

Colorectal cancer

Refractory, primary or

metastatic liver cancer

Paediatric cancers including

neuroblastoma, hepatocellular

carcinoma and Ewing sarcoma

Phase I

Phase I

Phase 1b

[95-97]

NDV ATV-NDV Non-engineered, stimulates anti-cancer

immune response.

IT

Colorectal, breast, glioblastoma

and SCCHN

Pilot studies

[71-73]

VSV Recombinant VSV Gene insertion for IFN-β production. IT Refractory or intolerant

hepatocellular carcinoma Phase I [78]

CVA CVA-21

Non-engineered, targets ICAM-1 and

DAF receptors on surface of tumour cells.

Stimulates immunogenic tumour

cytotoxicity and priming of an adaptive

immune response.

IT Metastatic melanoma Phase II [85, 86]

16

1.3. ONCOLYTIC REOVIRUS

1.3.1. dsRNA genome and molecular structure

Reoviruses belong to the orthoreovirus genus of the reoviridae family [98]. They are

commonly found in un-treated sewage and stagnant water, and infect a wide range of

hosts including horses, cattle, cats, dogs, mice, birds, crustaceans and humans. These

viruses were isolated from the respiratory or enteric tracts of children during the

1950’s who had been hospitalised with diarrheal illnesses [99]. Reoviruses can

occasionally cause mild gastrointestinal or flu-like symptoms, but are not associated

with serious human disease and are therefore known as orphan viruses [100]. Most

people have been exposed to reovirus infection during infancy, and approximately 50-

60% of the adult population test positive for reovirus-specific antibodies [101]. There

are three serotypes and four subtypes of mammalian reoviruses, type 1 Lang (T1L),

type 2 Jones (T2J), type 3 Dearing (T3D), and type 3 Abney (T3A) [99, 102].

Reoviruses are non-enveloped, but contain an inner and outer capsid that surrounds

the dsRNA genome. There are ten linear dsRNA segments, consisting of three large

(L), three medium (M), and four small (S) segments, which encode eleven viral

proteins (Table 1.4) [99]. Eight of the eleven primary translation products are

structural proteins that are present in mature reovirus particles, called virions, as

depicted in Figure 1.3. The other three proteins, namely µNS, σNS and σ1s, are non-

structural and have roles in reovirus replication.

Table 1.4. The 11 viral proteins encoded by the 10 dsRNA segments of the reovirus genome [99].

Gene Encoded protein(s) Protein function / property

L1 𝜆3 RNA-dependent RNA polymerase

L2 𝜆2 Guanylyltransferase and possible methyltransferase

L3 𝜆1 Binds dsRNA, zinc metalloprotein

M1 µ2 Associates with and stabilises cellular microtubules

M2 µ1 N-myristoylated, cleaved into fragments, role in

penetration and transcriptase activation

M3 µNS Binds ssRNA, associates with cytoskeleton, role in

assortment and secondary transcription

S1 σ1 + σ1s σ1 is the cell attachment protein

S2 σ2 Binds dsRNA

S3 σNS Binds ssRNA, role in assortment

S4 σ3 Sensitive to protease degradation, binds dsRNA,

zinc metalloprotein, effects on translation

17

Figure 1.3. Molecular structure of the reovirus virion. Image adapted from [102].

1.3.2. Replication life cycle

The first step in the replication cycle is the attachment of the virus to host cell surface

receptors, which is mediated by the viral σ1 protein [99, 103-105]. The σ1 protein

initially binds to sialic acid (N-acetylneuraminic) on the cell surface with low affinity

[106-108], before making contact with the junction adhesion molecule-A (JAM-A)

receptor with high affinity [109, 110]. Reovirus particles are then internalised by

receptor β1-integrin-mediated endocytosis [111], which occurs in a clathrin-dependent

manner [112]. Once inside the cell, the virions are surrounded within vacuoles that

resemble endosomes, where they are converted into infectious subvirion particles

(ISVPs). This occurs by degradation of the outer capsid σ3 protein by endosomal

proteases, conversion of protein µ1C to δ (its stable cleavage product), and

conformational changes in σ1 [113, 114]. Reovirus particles are therefore activated

through proteolysis, an acid dependent step [115]. Proteolytic activation of virions

can also occur extracellularly to generate ISVPs that directly penetrate the membrane,

thus bypassing clathrin-dependent endocytosis [99, 116, 117]. In order to form the

transcriptionally active core particle, the ISVP needs to be further processed.

Conformational rearrangements in the µ1 protein expose its myristoylated N-

terminus, before being auto-cleaved to form µ1N. µ1N then interacts with membrane

18

lipids to form pores that mediate the release of the viral core into the cytoplasm [118-

121]. Following this step, the core particle becomes transcriptionally active, and

contains all the necessary enzymes for primary transcription of the ten capped (at the

5’ end) plus-sense mRNA strands [98]. The plus-sense strands exit from the core into

the cytoplasm through channels in the λ2 core spike pentons, and are then translated

into viral proteins by the cellular protein synthesis machinery [99]. Early transcripts

associate with newly made viral proteins to form viral inclusions, where the plus-

strands serve as templates for minus-strand synthesis [98, 122, 123]. Secondary

transcription may also take place in the newly formed viral particles, which forms late

transcripts that serve as primary templates for viral protein synthesis later in infection

[99]. Upon assembly of the outer capsid, virions are released after cell lysis and may

subsequently infect adjacent cells [98]. This process is summarised in Figure 1.4.

Figure 1.4. The reovirus replication cycle [98].

19

1.3.3. Mechanism of selective replication in cancer cells

It was first noted in 1977 that transformed cell lines were more susceptible to

reovirus-induced cell death than un-transformed cell lines [124]. Similarly, data

published the following year showed that human lung fibroblast cells transformed

with the simian virus 40 (SV40) T-antigen became more sensitive to reovirus

oncolysis [125]. Since then, most of the research exploring the oncolytic potential of

reovirus has been based on the T3D subtype. This has resulted in the development of

Reolysin® (Oncolytics Biotech Inc); a naturally occurring, unmodified, replication

competent formulation of human reovirus T3D. Reolysin® has been used extensively

in clinical testing for the treatment of different cancers (Table 1.6). The two main

differences between the four reovirus subtypes are their abilities to agglutinate bovine

erythrocytes and their cell attachment protein sequences [99]. For example, T1L and

T3D share only 25% identity in their σ1 protein, but the outer capsid and viral core

proteins are highly conserved, showing 90-98% identity [126]. Data exploring the

oncolytic ability of the other subtypes is scarce, although Alloussi et al found that all

reovirus subtypes were able to lyse human glioma cell lines [127]. Reovirus T3D

(Reolysin®) is the main focus of this thesis.

1.3.3.1. Targeting of an aberrant Ras signalling pathway

Ras proteins are small GTP-binding hydrolysing proteins (GTPases) that exist in two

conformational states; GTP-bound active and GDP-bound inactive. Extracellular

signals through cell membrane receptors cause guanine nucleotide exchange factors

(GEFs) to replace GDP for GTP-bound Ras. Activated Ras then binds to a variety of

downstream effector molecules to stimulate the regulation of cellular proliferation,

differentiation and survival. Hydrolysis of GTP by Ras is facilitated by GTPase-

activating proteins (GAPs) [128, 129]. Mutations in the Ras gene renders its protein

product unresponsive to GAPs, chronically GTP-bound and active [129]. Activating

mutations in Ras genes have been found in >30% of all human cancers (not including

mutations in upstream activators or downstream effectors of Ras), thus promoting

angiogenesis, metastasis and loss of growth control [130, 131].

Several reports have shown that reovirus selectively replicates in cells with an

activated Ras pathway. Two reovirus-resistant mouse cell lines became highly

susceptible to the virus after transfection with the gene encoding EGFR, suggesting

20

that reovirus takes advantage of a functional EGFR signalling pathway for sufficient

oncolysis [132]. Further supporting this finding, the NIH-3T3 fibroblast cell line, that

is naturally resistant to reovirus and exhibits low EGFR expression, became highly

sensitive to reovirus-mediated cell death after transformation with the v-erbB

oncogene. This gene possesses ligand-independent, constitutive tyrosine kinase

activity, and encodes a protein structurally related to the EGFR [133]. Downstream

elements of Ras such as RalGEF and p38 have also been shown to mediate reovirus

oncolysis in NIH-3T3 cells [134]. It is thought that Ras transformed cells fail to

activate PKR after reovirus infection. PKR is a serine/threonine kinase that plays a

role in the antiviral interferon (IFN) response. In normal cells, PKR is activated via

trans-auto-phosphorylation in the presence of viral dsRNA. Activated PKR then

phosphorylates eIF-2α, resulting in an increased affinity for eIF-2B (a guanine

nucleotide exchange factor) and prevention of the exchange of GDP for GTP. Viral

protein synthesis is inhibited, as phosphorylated GDP-bound eIF-2α cannot partake in

the formation of the 43S pre-initiation complex [102]. Conversely, in cells with an

aberrant Ras signalling pathway, Ras, or one of its downstream elements, prevents the

phosphorylation of PKR. Thus, PKR remains in an inactivated state and cannot

terminate viral translation, allowing reovirus replication to proceed. This ultimately

leads to cell death [135]. This is demonstrated in Figure 1.5.

Figure 1.5. The proposed ‘reovirus-Ras’ model of selective oncolysis. Adapted from [131].

21

1.3.3.2. Studies that conflict the involvement of aberrant Ras signalling

Based on the reovirus-Ras model of selective oncolysis, it would seem logical to use

reovirus in tumours with a high occurrence of Ras mutations. However, a possible

limitation to the studies that investigated the involvement of the Ras pathway is that

they were based on experiments using murine fibroblasts rather than cancerous cell

lines. Later publications did show that the susceptibility of human pancreatic,

malignant glioma, and colon cancer cell lines (that often harbour the KRAS mutation)

to reovirus induced-cell death was associated with Ras activity [136-138]. However,

it has since been reported that active Ras signalling alone does not control the

susceptibility to reovirus oncolysis [139], as demonstrated in colon cancer cells and

non-small cell lung cancer (NSCLC) cell lines [140, 141]. Protease mediated-

disassembly of the reovirus outer capsid was shown to be a key determinant of

reovirus oncolysis that was independent of increased Ras activity [142, 143].

Terasawa et al found no correlation between Ras activation status and reovirus-

mediated cell killing in a number of different human tumour cell lines [144].

Likewise, Twigger et al carried out an extensive investigation on a panel of human

SCCHN cell lines with diverse sensitivities to reovirus oncolysis, but found no

significant association with Ras activation status or with the active signalling

component of EGFR [145]. They found that reovirus cytotoxicity did not depend on

major signalling pathways downstream of Ras, including MAPK, PI3K and p38, and

that PKR activation did not control the oncolytic effect of reovirus in SCCHN cell

lines [145]. As a result of these findings, Ras is not used as a routine biomarker of

reovirus sensitivity to a patient’s tumour in the clinic. Many cancer cell lines are

highly sensitive to reovirus oncolysis compared to normal cells. However, some are

surprisingly resistant, as demonstrated with certain SCCHN cell lines [145], and this

may reflect on the fact that a sub-population of patient’s tumours may not be as

responsive to reovirus therapy.

Ras signalling likely has an involvement in reovirus oncolysis in some cancer types,

but it is not the sole contributor. The full mechanism of reovirus-induced cancer cell

death is clearly much more complex and still remains to be elucidated. If this process

can be better understood, then biomarkers may be used to select only the patients

whose tumours will be responsive to reovirus treatment, and hence improve clinical

trial design. This would be of particular interest in the context of SCCHN, as this

22

cancer type has reached Phase III clinical status (the REO 018 trial), and it is not

certain as to whether the trial has reached its primary endpoint.

1.3.4. Biomarkers of treatment response

The field of oncology has entered an era of personalised medicine, where each cancer

patient’s treatment is customised according to their specific disease [146]. The

National Cancer Institute defines a biomarker as ‘a biological molecule found in

blood, other body fluids, or tissues that is a sign of a normal or abnormal process, or

of a condition or disease’ [147]. Biomarkers may be used to help distinguish the most

responsive patient subgroup to a cancer treatment [148], and thus have the potential to

lower costs and time. This type of marker is termed a predictive biomarker of

treatment response, and those that are in current use in the UK are displayed in Table

1.5. For example, KRAS is a predictive biomarker in tumour biopsy tissue of patients

with colorectal cancer, as somatic mutations in KRAS are associated with poor

response to EGFR-based therapies [149]. In breast and gastric tumours, gene

amplification or over-expression of the HER2 gene predicts for resistance to anti-Her2

agents, including trastuzumab [149]. Likewise, over-expression of the oestrogen

receptor in breast cancer is a marker of resistance to anti-endocrine therapies such as

tamoxifen [149].

To date, the only suggested biomarker of reovirus treatment response are cathepsins B

and L, which are proteases that are involved in the proteolytic disassembly of the

outer reovirus capsid proteins during infection. The activity level of these proteases

was shown to be higher in human tumour cell lines that are more susceptible to

reovirus-mediated oncolysis [144]. Although this study is encouraging, further

research is needed to establish host-cell factors that may be important in predicting the

anti-cancer effect after reovirus treatment, which is the purpose of Chapters 3, 4, and

5. In fact, the entire OV field lacks the use of such biomarkers, and an extensive

search of the literature revealed only two related publications. Flak et al used a

microarray approach to identify host cell factors that can influence the efficacy of

oncolytic adenovirus dl922-947 in ovarian cancer cell lines. They found that the

cyclin-dependent kinase inhibitor, p21, may be an important biomarker of treatment

response in clinical trials [150]. In another study, a microarray revealed that down-

regulation of immunoglobulin-like transcript 2 (ILT2) is a predictive biomarker of

23

clinical response in patients treated with oncolytic vaccinia virus B7.1 immunotherapy

[151].

Table 1.5. Tumour markers of treatment response currently in use in the UK. Adapted from [148].

Tumour marker Cancer type(s) Tissue analysed

ALK gene re-arrangements and over-

expression

Non-small cell lung cancer, anaplastic large cell

lymphoma Tumour

Alpha-fetoprotein (AFP) Germ cell tumours Blood

Beta-2-microglobulin (β2M) Multiple myeloma, chronic lymphocytic

leukaemia

Blood, urine,

cerebrospinal fluid

Beta-human chorionic gonadotropin

(β-hCG) Choriocarcinoma, germ cell tumours Urine, blood

BRCA1 and BRCA2 gene mutations Ovarian cancer Blood

BCR-ABL fusion gene (Philadelphia

chromosome)

Chronic myeloid leukaemia, acute lymphoblastic

leukaemia, acute myelogenous leukaemia

Blood and/or bone

marrow

BRAF V600 mutations Cutaneous melanoma, colorectal cancer Tumour

C-kit/CD117 Gastrointestinal stromal tumour, mucosal

melanoma Tumour

CA-125 Ovarian cancer Blood

CD20 Non-Hodgkin-lymphoma Blood

Chromogranin A (CgA) Neuroendocrine tumours Blood

EGFR gene mutation analysis Non-small cell lung cancer Tumour

Oestrogen receptor/progesterone

receptor Breast cancer Tumour

Fibrin/fibrinogen Bladder cancer Urine

HE4 Ovarian cancer Blood

HER2/neu gene amplification or

protein over-expression

Breast cancer, gastric cancer, gastroesophageal

junction adenocarcinoma Tumour

Immunoglobulins Multiple myeloma, Waldenström

macroglobulinemia Blood and urine

KRAS gene mutation analysis Colorectal cancer, non-small cell lung cancer Tumour

Lactate dehydrogenase Germ cell tumours, lymphoma, leukaemia,

melanoma, neuroblastoma Blood

Neuron-specific enolase (NSE) Small cell lung cancer, neuroblastoma Blood

Nuclear matrix protein 22 Bladder cancer Urine

Programmed death ligand 1 (PDL-1) Non-small cell lung cancer Tumour

Prostate specific antigen (PSA) Prostate cancer Blood

1.3.5. Pre-clinical testing of oncolytic reovirus T3D

Pre-clinical testing of reovirus T3D (Reolysin®) as a monotherapy, has shown anti-

cancer activity in a variety of different tumour cell lines, ex vivo specimens, and

murine SCID/NOD xenograft models [152, 153]. These pioneering studies

investigated reovirus oncolysis in cancers of the breast [154], brain [137], colon [138],

ovary [138], prostate [155], bladder [156], pancreas [136], and lung [157], and not to

mention various different blood cancers [152, 158, 159].

24

When combined with either chemotherapy or radiation, reovirus has demonstrated an

enhanced anti-tumour effect. For example, it was speculated that reovirus may target

radiation-resistant cell populations in tumours. This is because reovirus may replicate

in some cells with a constitutively activated Ras pathway, and resistance to radiation

in vitro is associated with over-expression of EGFR [160], activating mutations of Ras

[161], phosphorylation of Akt [162] and expression of PI3K [163]. Indeed, Twigger

et al found that combining reovirus with radiotherapy synergistically enhanced

cytotoxicity in a panel of tumour cells in vitro and in vivo [164]. Co-administration of

reovirus and the nucleoside analogue gemcitabine was more effective at killing human

colorectal cell lines than when the agents were used individually [152]. Synergistic

cell kill was evident in in vitro and in vivo malignant melanoma models after

treatment with reovirus and the chemotherapy drug cisplatin [165].

1.3.6. Clinical trials involving oncolytic reovirus T3D

Table 1.6 summarises the objectives and results of the 36 clinical trials involving

Reolysin® that have either been completed or are currently on-going. Phase I clinical

trials have demonstrated reovirus to be safe when administered systemically or

intratumourally. The maximum tolerated dose (MTD) has never been achieved and at

most, virus infection has caused fever, fatigue, muscle pain and headache [153]. As a

monotherapy, reovirus has exhibited modest anti-cancer activity in a variety of

different solid tumours, including glioma, PCa, metastatic melanoma, colorectal

cancer, and metastatic ovarian, peritoneal and fallopian tube cancers [166]. An

ongoing translational study is also assessing the safety and overall response rate of

patients with the haematological malignancy, multiple myeloma, after intravenous

reovirus treatment [166]. However, activation of innate and adaptive immune

responses can rapidly clear the virus and subsequently hinder its efficacy [153].

Hence, other clinical studies have combined reovirus with conventional treatments

such as radiation or chemotherapy, in an attempt to achieve an anti-cancer synergistic

effect.

25

Table 1.6. A summary of the Reolysin® clinical trials that have been completed or are currently ongoing. Monotherapy trials involving treatment only with Reolysin® are highlighted

in blue, whereas combination therapy trials are highlighted in orange. MTD = maximum tolerated dose. DLT = dose limiting toxicity. PR = partial response. Sd = stable disease. CR =

complete response. MED = minimum effective dose. PD = progressive disease. Adapted from [166].

Trial Number and Name

Phase,

Status, and

Location

Objectives of study Results of study Ref

REO 001: Local monotherapy of REOLYSIN® for patients with

subcutaneous tumours

Phase I

Complete

Canada

To determine the safety (DLT) and MTD of REOLYSIN® in patients with advanced solid tumours, who had otherwise

failed to improve on standard interventions.

None of the 18 terminal cancer patients receiving REOLYSIN® intralesionally experienced any serious adverse events, nor were there any DLTs. The MTD was not reached even at a dose of 1010 PFU. No viral

shedding was observed.

[167]

REO 002: Local monotherapy of

REOLYSIN® for patients with T2

prostate cancer

Translational

Complete

Canada

To examine the safety and the histopathological efficacy of

intratumoral administration of REOLYSIN® for the treatment

PCa restricted to the prostate gland.

Evidence of apoptotic tumour cell death in 4 of 6 patients after REOLYSIN® injection into the prostate gland 3

weeks prior to surgical removal of the prostate, with no safety concerns. CD8 T-cell infiltration within reovirus injected areas was also observed. One patient’s PSA levels dropped by 53% and observed prostate shrinkage

of 67% during the 3 week period prior to surgical removal.

[155]

REO 003: Local monotherapy of

REOLYSIN® for Patients with recurrent malignant gliomas

Phase I/II

Complete

Canada

To determine the MTD, DLT, and safety of REOLYSIN® when delivered intratumorally to patients with malignant

glioma. The secondary objective was to examine antitumor

activity.

12 patients were treated with a single injection at dosages of 1 x 107 TCID50, 1 x 108 TCID50 and 1 x 109 TCID50

in a delivery volume of 0.9 mL. MTD was not reached and REOLYSIN® was well tolerated. 3 patients lived

longer than 1 year, and 1 patient was still alive approximately 45 months post treatment. For the group as a

whole, the medium overall survival was 21 weeks.

[168]

REO 004: Systemic administration

of REOLYSIN® for patients with

metastatic tumours

Phase I

Complete

US

To determine the MTD, DLT and safety of REOLYSIN® in

patients with advanced or metastatic solid tumours. Secondary

objectives were to evaluate viral replication, immune

responses and antitumor activity.

18 patients were administered IV single doses of 1x108 or multiple doses every 4 weeks of 3x1010 TCID50. 8 patients showed Sd. A patient with progressive breast cancer experienced a 34% shrinkage in tumour volume.

Treatment was well tolerated and toxicities were mild (chills, fever, fatigue). All patients developed

neutralising antibodies. Viral shedding was observed in 6 patients and was associated with higher clinical benefit rate. Overall clinical benefit rate was 45%.

REO 005: Systemic administration


metastatic tumours

Phase I

Complete

UK

To determine the safety of REOLYSIN® when administered

intravenously. The secondary objective was to observe anti-

tumour and immune responses.

Patients were entered into the trial at a range of dose levels and treated to a maximum daily dose of 1 x 1011

TCID50. MTD was not reached and the treatment was well tolerated by the patients (grade 1 to 2 toxicities), with notable changes in stabilisation of disease, in addition to some minor tumour regressions in patients who

had failed all previous treatments.

[169,

170]

REO 006: Local administration of

REOLYSIN® in combination with

radiation for patients with advanced cancers

Phase I

Complete

UK

To determine the MTD, DLT and safety of REOLYSIN®

when administered intratumorally to patients receiving

radiation treatment. A secondary objective was to examine any evidence of antitumor activity.

23 patients with various solid tumours received 2 to 6 intratumoral doses of REOLYSIN® at escalating

dosages up to a maximum of 1 x 1010 TCID50 with a constant localised radiation dose of 20 Gy or 36 Gy. 2 patients in the low-dose (20 Gy) radiation group had a PR and 5 had Sd. 5 patients in the high-dose (36 Gy)

radiation group had PRs and 2 had Sd, for a clinical benefit rate (PR + Sd + CR) of 100%. The treatment was tolerated well in all cohorts, with no DLT, and the MTD was not reached. No viral shedding was observed.

[171]

REO 007: Infusion of REOLYSIN® for patients with recurrent malignant

gliomas

Phase I/II

Complete

US

To determine the MTD, DLT and safety of REOLYSIN®.

Secondary objectives included the evaluation of viral

replication, immune response and evidence of antitumor activity.

A single dose of REOLYSIN® was administered by infusion to patients with recurrent malignant gliomas that

were refractory to standard therapy. The phase I portion of the trial treated 5 patients in 5 cohorts with doses

escalating from 1x108 TCID50 to 1x1010 TCID50. The treatment was shown to be safe and well tolerated and MTD was not reached.

REO 008: Intratumoral

Administration of REOLYSIN® plus

low-dose radiation for patients with advanced malignancies

Phase II

Complete

UK

To assess the antitumor activity of REOLYSIN® plus low

dose radiotherapy in treated and untreated lesions. Secondary

objectives were to evaluate viral replication, immune response, safety, and tolerability.

16 patients who were heavily pre-treated with chemotherapy or radiation were enrolled in the trial. Of the 14

patients evaluable for response, 13 had Sd or better in the treated target lesions. Of these, PRs were observed in 4 patients (2 with melanoma and 1 each with lung and gastric cancer) and minor responses were observed in 2

patients (thyroid, ovarian), for a total disease control rate (Sd + PR + CR) of 93% in the treated lesions.

Treatment was well tolerated.

[172]

REO 009: Intravenous administration of REOLYSIN® plus gemcitabine for

patients with advanced malignancies

Phase I

Complete

UK

To determine the MTD, DLT and safety. Secondary objectives

were to evaluate the immune response to the drug combination

compared to chemotherapy alone and evidence of antitumor activity.

16 patients were enrolled. 2 initial patients treated with up to 3x1010 TCID50 reovirus, and 1 patient in the final cohort, experienced DLTs due to a grade 3 rise in liver enzymes, likely caused by concomitant aminocetophen

use. The liver enzyme increase was however transitory and reversible. Reovirus dose was lowered to 1x109

TCID50. No viral shedding was observed. Of the 10 patients evaluable for response, 2 patients (breast and nasopharyngeal) had PRs and/or clinical response, and 5 patients had Sd for between 4 and 8 cycles, for a total

disease control rate (CR + PR + Sd) of 70%.

[173]

26

REO 010: Intravenous administration of REOLYSIN® plus docetaxel for

patients with advanced malignancies

Phase I

Complete

UK

To determine the MTD, DLT, recommended dose, dosing

schedule and safety. Secondary objectives were to evaluate

immune response compared to chemotherapy alone and any evidence of antitumor activity.

25 patients were enrolled, with 24 being exposed to treatment and 23 completing at least 1 cycle of therapy. 16

patients were suitable for response assessment. Combination of docetaxel (75mg/m2on day 1) and reovirus (escalating doses up to 3x1010 TCID50 on days 1-5) every 3 weeks, was deemed to be safe and well tolerated,

and MTD was not reached. Antitumor activity was seen, with one complete response and three PRs. A disease

control rate (CR +PR + Sd) of 88% was observed.

[174]

REO 011: Intravenous administration of REOLYSIN® plus paclitaxel and

carboplatin for patients with advanced

head and neck cancers

Phase I/II

Complete

UK

To determine MTD, DLT, recommended dose, dosing schedule and safety. Secondary objectives were to evaluate

immune response compared to chemotherapy alone and any

evidence of antitumor activity.

Triple therapy of REOLYSIN®, paclitaxel and carboplatin was well tolerated when administered intravenously. Of 19 evaluable patients with SCCHN refractory to prior platinum-based chemotherapy for

recurrent, metastatic disease, 8 experienced PRs and 6 had Sd. The total clinical benefit rate (CR + PR + Sd)

was 74%.

[175]

REO 012: Intravenous administration of REOLYSIN® plus

cyclophosphamide for patients with

advanced malignancies

Phase I

Complete

UK

To determine MED of cyclophosphamide for successful

immune modulation. Secondary objectives were to assess safety and evidence of antitumor activity.

Patients received REOLYSIN® on days 1 to 5 of each 28 day treatment cycle, with cyclophosphamide 3 days

prior to the start of the first cycle and then on day 26 of each cycle, in the absence of DLTs. 30 patients were

enrolled. Treatment combination was safe, with Grade 3 or 4 toxicities seen only in patients at or above the MTD of cyclophosphamide. Thus, cyclophosphamide did not attenuate host antiviral responses, but

association with PBMCs may allow reovirus to persist and evade even high levels of neutralizing antibodies.

[176]

NCI 7853: Systemic and intraperitoneal administration of

REOLYSIN® for patients with

metastatic ovarian, peritoneal and fallopian tube cancers

Phase I

Complete

US The primary objectives were safety, tolerability and MTD.

Patients received a constant dose of intravenous REOLYSIN® on days 1 to 5 of each 28 day cycle, as well as

an escalating dose of intraperitoneal REOLYSIN® on days 1 and 2. 14 patients were enrolled. Treatment was

safe and well tolerated, with no DLTs observed.

[177]

NCI-7848: Intravenous administration of REOLYSIN® for

patients with metastatic melanoma

Phase II

Complete

US

Assess antitumor effects of REOLYSIN®, as well as its safety profile. Secondary objectives included assessment of

progression free survival and overall survival.

21 patients received systemic administration of REOLYSIN® at a dose of 3 x 1010 TCID50 per day on days 1 to

5 of each 28 day cycle, for up to 12 cycles of treatment. Treatment was well tolerated. Median time to

progression-free and overall survival were 45 days and 165 days respectively. Viral replication was demonstrated in biopsy samples.

[178]

REO 013: Intravenous administration


metastatic colorectal cancer

Translational

Complete

UK

Assess presence, replication and anti-cancer effects of reovirus within liver metastases after intravenous administration of

REOLYSIN®. Secondary objectives were to assess antitumor

activity, safety, and monitoring the humoral and cellular immune responses.

REOLYSIN® was given for 5 consecutive days in advance of surgery to remove colorectal cancer deposits

metastatic to the liver. Patients comprised 2 groups receiving REOLYSIN®, either at an early (10 to 21 days) or late (less than 10 days) time point before surgical resection. All patients had pre-existing immunity to the

virus, but reovirus could still target and infect metastatic liver tumours in 90% of the patients. Reovirus was

able to evade these neutralizing effects of the immune system by binding to specific blood cells that in turn delivered the virus to the tumour. Analysis of surgical specimens demonstrated greater expression of reovirus

protein in malignant cells compared to either tumour stroma or surrounding normal liver tissue.

[179]

REO 014: Intravenous administration of REOLYSIN® for patients with

metastatic sarcomas

Phase II

Complete

US

To measure tumour responses, duration of response, and evidence of antitumor activity after multiple-dose

REOLYSIN® treatment.

Treatment was well tolerated. 19 of 44 evaluable patients experienced Sd ranging from 2 to 22 months,

resulting in a total clinical benefit rate of 43%. The response objective for the study was 3 or more patients having prolonged stabilisation of disease (at least 6 months) or better, in order for the agent to be considered.

The trial exceeded its established objective with 6 patients experiencing Sd for more than 6 months. At the time

of reporting, 2 patients had experienced Sd for more than 19 months.


carboplatin for patients with advanced

head and neck cancers

Phase II

Complete

US

Measure duration of response and evidence of antitumor

activity. The secondary objective were to determine the safety and tolerability.

This was a confirmatory study using same combination treatment as the REO 011 trial. Of the 14 enrolled

patients, all had received prior chemotherapy, radiotherapy, or combinations thereof for their metastatic or

recurrent disease. 10 of the 14 patients received prior chemotherapy treatment with taxanes. Of the 13 patients evaluable for response, 4 had PRs, for an objective response rate of 31%. 6 patients had Sd or better for 12

weeks or longer for a disease control rate (stable disease or better) of 46%.


of REOLYSIN® plus paclitaxel and carboplatin for patients with non-

small cell lung cancer

Phase II

Complete

US

Determine the objective response rate in patients with KRAS or EGFR-activated tumours, and to measure progression-free

survival at 6 months. Secondary objectives were median

survival, duration of progression-free survival, safety and tolerability.

Patients received paclitaxel and carboplatin on day 1 of each 21 day cycle, with REOLYSIN® administered on

days 1 to 5. 37 patients were enrolled. Median progression-free survival for the study was 4 months, and median overall survival was 13.1 months. Of the 35 patients evaluable for clinical response, 11 patients

(5 KRAS mutant) had a PR, 20 had Sd and 4 had PD, for an objective response rate of 31%.


of REOLYSIN® plus gemcitabine for

patients with advanced pancreatic

cancer

Phase II

Complete

US

Determine the clinical benefit rate of treatment. The

secondary objectives were to determine the progression-free survival, safety and tolerability.

Patients received gemcitabine on days 1 and 8 of each 21 day cycle, and REOLYSIN® on days 1, 2, 8 and 9.

33 patients were enrolled in the study. Median progression-free survival for the study was 4 months, and

median overall survival was 10.2 months. The data suggested that the drug combination resulted in approximately 2-fold increase in one-year survival rates. Of the 29 patients evaluable for clinical response, 1

patient had a PR, 23 had Sd and 5 had PD. This translated into a clinical benefit rate of 83%.

27


of REOLYSIN® in combination with

paclitaxel and carboplatin for patients with platinum-refractory head and

neck cancers

Phase III

Complete

International

A double-blinded, randomized, 2-arm study assessing triple therapy versus chemotherapy alone. The primary endpoint

was overall survival, and secondary objectives were

progression-free survival, objective response rate, safety, and best percentage tumour-specific response in loco-regional and

metastatic disease.

A total of 167 patients were enrolled into 2 groups: patients with local recurrent disease with or without distal

metastases, and those with only distal metastases. Patients received standard doses of paclitaxel and carboplatin on day 1 of each 21-day cycle, and on days 1 to 5 with either intravenous placebo or REOLYSIN® at 3x1010

TCID50. Test arm patients with loco-regional disease demonstrated a progression-free and overall survival

benefit over control arm patients through 5 cycles of therapy (p=0.0072 and p=0.0146 respectively). The 118 patients with loco-regional disease, with or without distal metastases, were evaluated for percentage magnitude

of tumour shrinkage at the first post-treatment scan. The test arm showed a statistical trend towards increased

tumour shrinkage over the control arm (p=0.076). REOLYSIN® was found to be safe. Patients on the test arm experienced a higher incidence of flu-like symptoms.


carboplatin for patients with

metastatic melanoma

Phase II

Complete

US

Measure the objective response rate of the treatment

combination. Secondary objectives included assessment of

progression-free and overall survival, as well as assessment of disease control rate, safety and tolerability of the combination

treatment.


days 1 to 5. Up to 43 patients were to be enrolled in the study: 18 evaluable patients were enrolled in the first

stage and, subject to meeting response endpoints, the remainder in the second stage. The endpoint was met after 14 evaluable patients were enrolled. 3 of 14 patients exhibited a PR, and an additional seven patients had

Sd for a disease control rate of 71.5%. Final data pending.


of REOLYSIN® plus paclitaxel and carboplatin for patients with

squamous cell carcinoma lung cancer

Phase II

Complete

US

Assess the antitumor effect in terms of objective response rate. Secondary objectives were to assess progression-free and

overall survival, the proportion of patients receiving the

treatment who are alive and free of disease progression at 6 months, and to assess safety.


days 1 to 5. 25 evaluable patients received between 2 and 12 cycles of therapy. Of these, 92% exhibited overall tumour shrinkage. When evaluated for overall response, 40% had PRs, 48% showed Sd and 12% had

PD. 31.8% of patients with sufficient follow up had progression-free survivals greater than 6 months.


of REOLYSIN® plus FOLFIRI for patients with colorectal cancer

Phase I

Ongoing

US To determine a MTD and DLT with the combination. Results pending.

GOG-0186H (NCI): Intravenous

administration of REOLYSIN® plus

paclitaxel for patients with persistent or recurrent ovarian, fallopian tube or

primary peritoneal cancer

Phase II

Ongoing

US

Primary objectives are progression-free survival and toxicity.

The secondary objectives are overall survival by treatment group, progression-free survival group, and tumour response

Results pending.

COG-ADVL1014 (NCI):

Intravenous administration of

REOLYSIN® plus cyclophosphamide

for paediatric patients with relapsed or refractory solid tumours

Phase I

Complete

US

Estimate MTD and toxicities of treatment. Secondary

objectives are to measure antitumor activity and neutralising

antibodies.

Patients received REOLYSIN® on days 1 through 5 of each 28 day cycle, with some patients also receiving

cyclophosphamide on days 1 through 21. 29 patients were enrolled. There were no hematologic DLTs. The

median time to clear reovirus viremia was 6.5 days.

[180]

NCI-8601: Intravenous administration of REOLYSIN® plus

paclitaxel and carboplatin for patients

with metastatic pancreatic cancer

Phase II

Ongoing

US

The primary objective is progression-free survival. Secondary

objectives include toxicity, overall response rate, overall

survival, measurement of immunologic markers, and

examination of whether a relationship may exist between Ras

pathway activation and response.

Results pending.

NCI-9030: Intravenous administration of REOLYSIN® in

patients with relapsed multiple

myeloma

Phase I

Complete

US

Primary objectives were toxicity and MTD. Secondary

objectives were duration of response, objective response rate, progression-free survival and time to progression.

Patients received REOLYSIN® on days 1 to 5 of each 28 day treatment cycle for up to 12 cycles in the absence

of disease progression or unacceptable toxicity. 12 patients were enrolled and the treatment was safe and well tolerated.

[181]

IND 209: Intravenous administration

of REOLYSIN® plus docetaxel in patients with recurrent or metastatic

castration resistant prostate cancer

Phase II

Ongoing

Canada

The primary objective is the efficacy of the treatment, based

on disease progression at 12 weeks. Secondary objectives

include the effect on circulating tumour cell status, objective response rate, overall survival, and the measurement of

molecular factors which may be prognostic or predictive of

response.

Results pending.

IND 210: Intravenous administration of REOLYSIN® plus FOLFOX-6 and

bevacizumab, versus FOLFOX-6 and

Phase II

Ongoing

Canada

The primary objective is progression-free survival. Secondary objectives include changes in CEA levels, objective response

rate, overall survival, quality of life, the tolerability and

Results pending.

28

bevacizumab alone in patients with

advanced or metastatic colorectal cancer

toxicity of the treatment combination, and measurement of

molecular factors which may be prognostic or predictive of response.


of REOLYSIN® plus docetaxel or

pemetrexed for patients with advanced or metastatic non-small cell

lung cancer

Phase II

Ongoing

Canada

Primary objective is progression-free survival. Secondary

objectives include tolerability and toxicity of treatment,

progression rates at 3 months, objective response rate, overall survival and measurement of molecular factors which may be

prognostic or predictive of response.

Results pending.


of REOLYSIN® plus paclitaxel for patients with advanced or metastatic

breast cancer

Phase II

Ongoing

Canada

The primary objective is progression-free survival. Secondary objectives are objective response rate, overall survival,

circulating tumour cell counts, toxicity of the treatment, and

measurement of molecular factors which may be prognostic or predictive of response.

Results pending.

REO 019: Intravenous

administration of REOLYSIN® plus bortezomib and dexamethasone in

patients with relapsed or refractory

multiple myeloma

Phase 1b

Ongoing

US

Primary objectives are to determine the safety, MTD, and

overall objective response rate. The secondary objectives

include objective response rate at escalating doses, and progression-free and overall survival.

Results pending.

REO 024: Intravenous administration of REOLYSIN® plus

pembrolizumab and chemotherapy in

patients with advanced or metastatic pancreatic adenocarcinoma

Phase 1b

Ongoing

US

Primary objectives are safety and DLT. Secondary objectives are overall response rate and progression-free survival by

immune-related response criteria, overall survival, and the

effects of the treatment combination by analysis of pre- and post-treatment biopsies and blood-based immune markers.

Results pending.

MAYO-(MC-1472): Intravenous

administration of REOLYSIN® plus GM-CSF in paediatric patients with

relapsed or refractory brain tumours

Phase 1

Ongoing

US

The primary objective is safety and tolerability. Secondary

objectives include median progression-free and overall

survival.

Results pending.

NCI-9603: Intravenous

administration of REOLYSIN® plus dexamethasone and carfilzomib for

patients with relapsed or refractory

myeloma

Translational

Ongoing

US

Primary objectives include measuring reovirus replication and

safety. Secondary objectives include examining objective response, duration of response, clinical benefit, progression-

free survival, time to progression, and the measurement of

immunologic correlative markers.

Results pending.

REO-013 Brain: Intravenous

administration of REOLYSIN® in

patients prior to surgical resection of

recurrent high grade primary or

metastatic brain tumours

Translational

Ongoing

UK

The primary objective is to assess the presence of reovirus

within specimens taken from resected brain tumours. The

secondary objectives are safety, humoral and cellular immune

response, and to assess the replication and antineoplastic

effects of reovirus.

Results pending.

29

1.3.7. Combining reovirus T3D with metronomic doses of taxane chemotherapy

drugs

Taxane chemotherapy drugs display anti-cancer properties by binding to

microtubules, the intracellular filaments that form part of the cell’s cytoskeleton.

Microtubules are composed of α-tubulin and β-tubulin heterodimers, and have a key

role in chromosome separation during cell division and mitosis [182]. The binding of

taxanes to tubulin promotes the stabilisation of GDP-bound tubulin in the microtubule

resulting in inhibition of disassembly and prevention of subsequent mitosis and cell

division [183], which ultimately leads to cell death. Taxane drugs selectively target

cancer cells as they are able to rapidly grow through continuous mitotic division, and

are therefore more sensitive to inhibition of mitosis than normal healthy cells. The

first generation taxane drug, Paclitaxel (trade name Taxol®), was isolated in the

1960’s from the bark of the pacific yew Taxus brevifolia, and was first approved in

1992 by the FDA for the treatment of ovarian cancer. Docetaxel (trade name

Taxotere®) is a second generation semisynthetic analogue of Paclitaxel, and was

shown to be more cytotoxic than Paclitaxel on proliferating tumour cells [184]. The

FDA first approved Docetaxel in 1996 for the treatment of advanced breast cancer,

and then for the treatment of metastatic CRPC in 2004 [182]. Both Paclitaxel and

Docetaxel show the ability to inhibit tumour growth and improve patient survival in

different cancer types [185]. However, the limitation of these drugs is the

development of resistance in cancer cells, which is mainly associated with increased

expression of the multidrug resistance protein-1 (MDR1) gene that encodes P-

glycoprotein [185]. P-glycoprotein is an ATP-dependent drug efflux pump which

decreases the concentration of these drugs in tumour cells. A newer semisynthetic

taxane drug called Cabazitaxel (trade name Jevtana®) was approved by the FDA in

2010 for the treatment of castration resistant metastatic PCa, in patients who had

previously been treated with a regimen containing Docetaxel [182]. Cabazitaxel has

poorer affinity to P-glycoprotein compared with Paclitaxel and Docetaxel, due to the

presence of extra methyl groups in its chemical structure [186]. It therefore represents

a promising antitumour therapeutic as it retains activity against Docetaxel-resistant

cancer cell lines. Cabazitaxel was also shown to have greater penetration of the

blood-brain barrier compared with Docetaxel and Paclitaxel, and stabilises

mictotubules against cold-induced depolymerisation in vitro as potently as Docetaxel

30

[187, 188]. Reovirus has been shown to associate with and stabilise microtubules by

the viral µ2 protein, and viral growth was reliant on µ2-mediated recruitment of viral

factories to microtubules [189-191]. The combination of two microtubule stabilisers

(reovirus and Docetaxel) promoted synergistic PCa cell death in vitro and reduced

tumour growth in vivo, in a PC3 xenograft murine model. This led to an increase in

apoptotic and necrotic cell populations, and the mechanism of synergy was partially

explained by an increase in microtubule stability, which was accompanied by an

increase in viral titre at early time points in cells treated with the combined therapy

[192]. This observation was also noted in an earlier study that treated NSCLC cell

lines with reovirus in combination with Paclitaxel [141].

Conventional chemotherapy treatment normally involves treating patients with the

maximum tolerated dose (MTD) in order to kill as many cancer cells as possible.

This is considered the highest concentration of drug that can be tolerated in regards to

toxicity and side-effects. Typically, the MTD would be given in 3 week cycles. A

cycle is the time between one round of treatment and the start of the next. There is a

drug-free break in-between each cycle of treatment to allow the patient to recover

from the harmful effects of the cytotoxic drugs. For example, a patient with CRPC

may be given Paclitaxel on day 1, and will then be drug-free until day 21 when the

cycle is repeated. It has been shown that shortening of the drug-free period between

each chemotherapy cycle limits the amount of tumour vasculature re-growth, thus

enhancing the anti-cancer effect [193]. The primary targets of conventional

chemotherapy regimens are the cancer cells, although unwanted side effects are also

commonly observed due to the cytotoxic drugs affecting all dividing cells, including

healthy normal cells [194]. Angiogenesis has been shown to be a key driver in the

growth of cancer [195]. Angiogenesis inhibitors have been developed, such as the

monoclonal antibody Bevacizumab (trade name Avastin®) that inhibits VEGF-A, a

protein that stimulates the growth of blood vessels. There has been a debate on how

to optimally use Bevacizumab and that using it continuously would be costly [196].

Metronomic chemotherapy (MC) is defined as the frequent administration of

chemotherapy agents at doses below the MTD and with no prolonged drug-free

breaks. It is believed that MC mainly targets endothelial cells involved in tumour

angiogenesis [197]. This may occur by directly killing circulating endothelial cells

(CEC) in the tumour vasculature or by killing bone-marrow-derived endothelial

31

progenitor cells (EPC) [195]. MC can also stimulate the immune system to initiate an

anti-tumour response [198-202]. There is limited Phase III clinical trial data on the

use of MC. However, effective anti-cancer responses and a better quality of life due

to the low toxicity of MC have been demonstrated in various pilot and phase II

clinical studies [203]. Cyclophosphamide, Methotrexate, Trofosfamide and Etoposide

are examples of agents that have been tested in a metronomic fashion in patients with

breast cancer, ovarian cancer, metastatic melanoma, prostate cancer, recurrent

glioblastoma and NSCLC [204-210]. MC also has the potential to be a relatively

inexpensive treatment [203], yet the manner in which MC is applied needs to be

further validated in order to achieve the best outcome. This includes finding the

minimally effective dose, the frequency of treatment and the choice of agent to use

[203]. Tubulin inhibitors may be effective in the metronomic setting [211, 212].

Paclitaxel was shown to selectively inhibit the proliferation of endothelial cells in

vitro at very low picomolar concentrations, much lower than the standard dose [213].

In addition, low dose Docetaxel was four times stronger than low dose Paclitaxel in

causing organic and functional damage of human endothelial cells in vitro, and also in

vivo by using the chick embryo chorioallantoic membrane (CAM) model [214]. MC

treatment of Cabazitaxel has not yet been explored, nor has the combination of MC

with oncolytic viruses. Although the MTD model is a convenient way to administer

drugs to patients, it is not necessarily the most beneficial way to use them clinically

[203]. As taxanes are used as a first line of care for PCa, it would be of particular

interest to determine whether an anti-cancer synergistic effect can be achieved with

the combination of reovirus and low doses of Cabazitaxel or Docetaxel, in

experimental models of PCa. This is addressed in Chapter 6.

32

Figure 1.6. A schematic representation of metronomic and conventional chemotherapy.

Metronomic chemotherapy (blue line) is administered in low doses below the maximum tolerated dose

(MTD) at regular intervals (weekly or daily), and aims to reduce the number of rapid proliferating

vascular endothelial cells (EC) or endothelial precursor cells (EPC) from the bone marrow that support

the growth of the tumour. Conventional chemotherapy (orange line) is given at high doses at the MTD,

and in cycles with relatively long drug-free breaks. Initially, conventional treatment kills many of the

cancer cells and some of the EC and EPC. However, the drug-free intervals allow the EC and EPC to

re-populate, which contributes to enhanced angiogenesis and eventual tumour re-growth.

33

1.4. SUMMARY

Reovirus T3D has demonstrated oncolytic activity across a wide range of

malignancies and has a relatively low toxicity profile. However, the use of reovirus

as an anti-cancer agent needs to be further optimised in order to progress through

clinical development.

Based on the fact that the mechanism of reovirus oncolysis is not fully understood,

and that there is no current biomarker of reovirus treatment response used in the

clinic, the objective of this research was to identify host-cell factors that may predict

for the susceptibility to reovirus-induced cell death in SCCHN cell lines. SCCHN is

of particular importance as this cancer type has entered Phase III clinical testing with

Reolysin®. If such a factor could be found, it may potentially help target the patients

whose tumours are most responsive to reovirus therapy, which may improve their

quality of life, as well as time, cost and trial outcome.

Conventional chemotherapy regimens at the maximum tolerated dose frequently cause

toxic side-effects and tumour vasculature re-growth. Given the paucity of data

describing the combination of reovirus and metronomic doses of chemotherapeutic

drugs, the work in the following chapters also attempts to determine whether the co-

administration of reovirus and low doses of taxane drugs achieves a synergistic anti-

cancer effect in PCa cell lines, compared to single-agent treatment. If in vitro models

of this system are successful, then this work may have a translational focus; it may

limit cytotoxicity and keep the number of endothelial cells that support tumour-

associated angiogenesis at bay, thus resulting in sustained clinical responses.

34

1.5. HYPOTHESIS AND OBJECTIVES

Firstly, we hypothesise that a predictive biomarker of reovirus treatment response

could be identified in SCCHN cell lines. In order to investigate this, the objectives of

this research are to:

Assess what effect the candidate genes, which were isolated by gene

expression profiling and microarray hybridisation, have on reovirus oncolysis

in SCCHN cell lines, using an siRNA-knock-down screen (Chapter 3).

Determine the effects of lipid-mediated over-expression or pharmacological

inhibition of any interesting target genes found in the siRNA screen, on

reovirus oncolysis (Chapter 4).

Study the biological mechanism of how these target genes influence the

reovirus oncolysis process in SCCHN cell lines (Chapter 5).

Compare the expression level of the corresponding target protein in head and

neck cancer and normal tissues (Chapter 5).

Secondly, we hypothesise that low, metronomic doses of taxane drugs in combination

with reovirus achieves a synergistic anti-cancer effect in PCa cell lines. To test this,

the objectives are to:

Determine the IC50 concentrations of reovirus, Cabazitaxel and Docetaxel in

PCa cell lines (Chapter 6).

Assess the interaction of reovirus and Cabazitaxel or reovirus and Docetaxel in

PCa cells at doses equal to or less than the IC50 values, by using two statistical

models. Concurrent and sequential dosing schedules will also be compared

(Chapter 6).

Study the biological mechanism of the interaction between the two agents

(Chapter 6).

35

CHAPTER 2

MATERIALS AND METHODS

36

2. MATERIALS AND METHODS

2.1. REOVIRUS AND CHEMOTHERAPEUTIC TAXANE DRUGS

Reovirus and the chemotherapeutic taxane drugs used in this study are shown in

Table 2.1, with details of their supplier and storage conditions. The reovirus stock

was titred via the plaque assay (Section 2.24) on the highly sensitive L929 cell line.

Table 2.1. Reovirus and Taxane drugs

2.2. CELL CULTURE MEDIA

All cell culture media used in this study are displayed in Table 2.2, along with their

supplier and supplements. Working cell culture media was prepared in the same way,

but was supplemented with 2% Fetal bovine serum (FBS).

Table 2.2. Cell Culture Media used in this study.

Reagent Supplier Storage conditions

Reovirus Type 3 Dearing

(T3D) (Reolysin®) Oncolytics Biotech Inc, Canada

Stored at -80°C in aliquots at

3×109pfu/mL

Cabazitaxel (Jevtana®) Produced by Sanofi Aventis, France. Kindly

donated by The Royal Surrey Hospital, UK.

Stored at -20°C in aliquots at

40mg/mL in ethanol

Docetaxel (Taxotere®) Sigma, UK Stored at -20°C in aliquots at

10mM in ethanol

Medium Supplements Supplier

Dulbecco's Modified

Eagle's Medium

(DMEM)

100 U/mL penicillin (Sigma, UK), 100µg/mL streptomycin (Sigma, UK),

2mM L-Glutamine (Sigma, UK) and 10% FBS (Life Technologies, UK). Sigma (UK)

Minimum Essential

Medium Eagle (MEM)



RPMI-1640 100 U/mL penicillin (Sigma, UK), 100µg/mL streptomycin (Sigma, UK),


F-12K 100 U/mL penicillin (Sigma, UK), 100µg/mL streptomycin (Sigma, UK),

2mM L-Glutamine (Sigma, UK) and 10% FBS (Life Technologies, UK).

American Type

Culture Collection

(ATCC, UK)

TRAMP-C2 DMEM


2mM L-Glutamine (Sigma, UK), 5% FBS (Life Technologies, UK), 5%

Nu-Serum IV (Corning, UK), 10nM dehydroisoandrosterone (Fisher

Scientific, UK) and 0.005mg/mL bovine insulin (Sigma, UK).

Sigma (UK)

37

2.3. CELL LINES

All cell lines (summarised in Table 2.3) were purchased from the American Type

Culture Collection (ATCC, USA), with the exception of PJ41 and PJ34 which were

obtained from the European Collection of Authenticated Cell Cultures (ECACC, UK),

and HN5 which was gifted to us by Professor Kevin Harrington at The Institute of

Cancer Research (ICR), London, UK [215]. The cells were adherent lines that were

maintained in a 37°C incubator in an atmosphere of 5% CO2, apart from the WPMY-1

cell line, which was maintained at 10% CO2. All tissue culture work was performed

in a sterile class II biosafety cabinet (Kendro, UK). Mycoplasma testing was carried

out regularly using a MycoAlert™ Mycoplasma Detection Kit (Lonza, UK). Cell

lines that were purchased prior to the start of this project were authenticated using

short tandem repeat (STR) profiling (LGC, USA), which confirmed ≥80% shared

alleles with a common ancestry.

Table 2.3. Cell lines used in this study, their growth media, tissue type and source.

Cell Line Cell culture medium

and supplements Tissue Type Source

PJ41 DMEM Human squamous cell carcinoma of the head

and neck ECACC, UK

HN5 DMEM Human squamous cell carcinoma of the head

and neck ICR, UK. [215]

PJ34 DMEM Human squamous cell carcinoma of the head

and neck ECACC, UK

MRC-5 MEM Human normal lung fibroblast ATCC, USA

COS-1 DMEM African green monkey kidney fibroblast ATCC, USA

HEK293A MEM Human embryonic kidney epithelial ATCC, USA

L929 DMEM Mouse fibroblast from subcutaneous

connective tissue; areolar and adipose ATCC, USA

DU145 MEM Human prostate carcinoma, derived from

brain metastasis ATCC, USA

PC3 F-12K Human prostate adenocarcinoma, derived

from bone metastasis ATCC, USA

LNCaP RPMI-1640 Human prostate carcinoma, derived from

lymph node metastasis ATCC, USA

WPMY-1 DMEM (with 5%

FBS) Human normal prostate stroma/fibroblasts ATCC, USA

TRAMP-C2 TRAMP-C2 DMEM Mouse transgenic prostate adenocarcinoma ATCC, USA

38

2.4. PASSAGING OF ADHERENT CELLS

Media was removed from the culture flask and the cell sheet was washed using

Hanks’ balanced salt solution (Sigma, UK) before discarding. Cells were detached

using Trypsin-EDTA (x1) solution (Sigma, UK) for 5-15 minutes at 37°C. When the

cells were completely detached, cell culture media was added to the flask to disperse

cells, which was then transferred to a universal tube. After centrifuging the tube at

1500rpm for 3 minutes, the supernatant was poured off to leave the cell pellet at the

bottom, which was re-suspended in 1mL cell culture media. To split the cells 1:4,

0.25mL of the cell suspension was added to a new T-75 flask containing 12.5mL fresh

cell culture medium and incubated at 37°C. To set up cells for an experiment, the re-

suspended cells were counted to enable seeding at an appropriate cell density.

2.5. EVALUATION OF CELL NUMBER

A 1:10 dilution of cell suspension in trypan blue (Sigma, UK) was made and 10µL of

the mixture was loaded onto the grid of a Neubauer haemocytometer. Trypan blue is

membrane impermeable and can only penetrate dead cells, and therefore viable cells

remain unstained. Only viable cells were counted in the four corner squares of the

haemocytometer. The following formula determined the cell number:

Mean number of cells per quadrant × dilution factor × 104 = number of cells/mL

2.6. CRYOPRESERVATION OF CELLS

Cells were harvested in cell culture medium and centrifuged at 1500rpm for 3

minutes. The supernatant was removed and the cell pellet was re-suspended to 1x106

to 1x107 cells/mL in freezing medium containing 50% cell culture medium, 40% FBS

(Life Technologies, UK) and 10% Dimethyl sulphoxide (DMSO) (Sigma, UK). The

cell suspension was transferred to sterile cryovials and stored at -80°C in a cryo-

freezing container, ensuring a controlled decrease in temperature, before being

transferred to liquid nitrogen for long-term storage.

39

2.7. REVITALISATION OF CRYOPRESERVED CELLS

Cryopreserved cells were removed from liquid nitrogen and thawed rapidly in a 37°C

waterbath for 2 minutes. Cells were then transferred to a universal tube containing

10mL warm cell culture medium and centrifuged at 1500rpm for 3 minutes. The

supernatant was poured off to remove traces of DMSO and the cell pellet was re-

suspended in 8mL cell culture media, which was transferred to a T-25 flask and

incubated at 37°C.

2.8. CALCULATING THE VOLUME OF REOVIRUS NEEDED FOR A

CERTAIN MULTIPLICITY OF INFECTION (MOI)

The definition of an MOI is the ratio of the number of infectious virus particles to the

number of target cells present in a defined space. However, the actual number of

virus particles that will enter any given cell is a statistical process, as some cells may

absorb more than one virus whereas others may not absorb any. The probability ‘P’

that a cell will absorb ‘n’ virus particles when inoculated with an MOI of ‘m’ can be

calculated for a given population using the Poisson distribution [99, 216]:

To calculate the volume of reovirus needed for a certain MOI, the following

calculations were performed:

1. number of cells/well × required MOI = pfu/well

2. (pfu/well ÷ 3.00×109pfu/mL reovirus stock) = mL/well reovirus needed

40

2.9. CELL TITRE 96® AQUEOUS NON-RADIOACTIVE CELL

PROLIFERATION (MTS) ASSAY

The Cell Titer 96® AQueous One Solution Reagent (Promega, UK) contains an MTS

tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-

(4-sulfophenyl)-2H-tetrazolium, inner salt], and an electron coupling reagent

(phenazine ethosulfate; PES). PES has enhanced chemical stability that allows it to

be combined with MTS to form a stable solution. The MTS compound is bio-reduced

by cellular oxidoreductase enzymes into a coloured formazan product (Figure 2.1),

which is directly proportional to the number of living cells in culture.

Cells were seeded (100µL per well) in 96-well plates at the required seeding density

in cell culture media for 24 hours at 37°C, to ensure an 80% confluence. 100µL

diluted reovirus (Section 2.8), chemotherapeutic drug or working media (un-treated

cells) was added to the wells of the plate in triplicate for an appropriate time-period at

37°C. 100µL Cell Titre 96® AQueous One Solution Reagent (Promega, UK), diluted

1:10 in RPMI working media, was added to each well, including empty wells for use

as a background control. The plates were incubated for 1-4 hours at 37°C, making

sure to incubate for the same duration within each individual experiment, before

reading the Optical Density (OD) absorbance readings on the Variskan® Flash plate

reader (Thermo Scientific, UK) at wavelength 492nm. The data was then analysed.

The average OD of the background control was subtracted from the average OD of

each sample. The % cell survival in each treatment was then calculated relative to the

un-treated cells by using the following formula:

% cell survival = (average OD treated sample ÷ average OD un-treated sample) × 100

The fraction of dead cells affected by reovirus or chemotherapeutic drug was then

calculated:

Fraction affected = 1 – (% cell survival ÷ 100)

The fraction affected at each dose was inputted into CalcuSyn software (Biosoft, UK)

to find the concentration of reovirus or drug needed to produce a 50% inhibitory

effect (IC50), using the median effect methods of Chou [217] (Section 2.27.4).

41

Figure 2.1. Chemical structures of the MTS tetrazolium compound and its formazan product. Image adapted from the CellTiter 96® AQueous Non-Radioactive Cell Proliferation Assay manual:

http://www.promega.co.uk/resources/protocols/technical-bulletins/0/celltiter-96-aqueous-

nonradioactive-cell-proliferation-assay-protocol/.

2.10. RNA EXTRACTION FROM CELL LINES

Total RNA was extracted from cell lines using the RNeasy® Plus Micro Kit (Qiagen,

UK), according to the manufacturer’s instructions. In summary, cells were lysed and

homogenised using β-mercaptoethanol (Sigma, UK) diluted 1:100 in Buffer RLT Plus

for 5 minutes at room temperature. The lysate was then passed through a gDNA

eliminator spin column to remove genomic DNA. Ethanol (Fisher Scientific, UK)

was added to the flow through to provide appropriate binding conditions for the RNA,

before transferring the sample to an RNeasy spin column. Contaminants were washed

away using specific buffers, leaving the RNA bound to the silica membrane of the

column. The RNA was eluted with 14µL RNase-free water and the concentration

(ng/µL) was analysed using the NanoDrop® ND-1000 Spectrophotometer (Labtech

International, UK). RNA samples were used immediately for cDNA synthesis or

stored at -80°C.

2.11. COMPLEMENTARY DNA (cDNA) SYNTHESIS FROM CELL LINES

cDNA was reverse transcribed from total RNA using the Cloned AMV First-Strand

cDNA Synthesis Kit (Life Technologies, UK), by following the manufacturer’s

instructions. Prior to cDNA synthesis, the RNA template and primer was denatured in

the absence of reaction buffer and enzyme to remove secondary structure that may

42

impede full-length cDNA synthesis. For each reaction, 1µL 50µM Oligo (dT)20

primer, 500ng RNA, and 2µL 10mM dNTP mix was combined, and the volume was

adjusted to 12µL with DEPC-treated water. The samples were incubated at 65°C for

5 minutes and then placed directly onto ice. Next, 4µL 5x cDNA Synthesis buffer,

1µL 0.1M DTT, 1µL 40U/µL RNaseOUT, 1µL DEPC-treated water, and 1µL

15U/µL Cloned AMV reverse transcriptase was added to each RNA reaction tube, to

give a total volume of 20µL. The reaction tubes were transferred to a pre-heated

thermal cycler (Applied Biosystems, UK), and incubated at 50°C for 1 hour. The

reaction was terminated by incubating at 85°C for 5 minutes. Assuming that the

cDNA synthesis was 100% efficient, 500ng cDNA was produced in 20µL to give a

concentration of 25ng/µL, which was further diluted to 5ng/µL in DEPC-treated

water. The cDNA’s were used immediately in the RT-qPCR reaction or stored at -

20°C.

2.12. REAL TIME–QUANTITATIVE POLYMERASE CHAIN REACTION

(RT-qPCR)

RT-qPCR analysis was performed using the Stratagene Mx3005P qPCR machine

(Agilent Technologies, USA), using SYBR Green fluorescence to measure the

amount of PCR product. A RT-qPCR master mix was prepared by adding 12.5µL

SYBR Green Jumpstart Taq Ready Mix (Sigma, UK), 0.25µL Reference Dye (ROX)

(Sigma, UK), 5.25µL RNase-free water and 2µL cDNA (5ng/µL), giving a total

volume of 20µL per reaction. 5µL target primer was added to the appropriate wells of

the 96-well plate containing the master mix, in duplicate. The house-keeping gene β-

actin was used as an endogenous control. The reaction conditions were: 1 cycle (10

minutes at 95°C), followed by 40 cycles (30 seconds at 95°C, 1 minute at 60°C and 1

minute at 72°C).

The forward and reverse primers for all human target genes used in the RT-qPCR

reaction are displayed in Table 2.4, and were designed by Professor Richard Morgan

using the Primer3 and BLAST tool (http://www.ncbi.nlm.nih.gov/tools/primer-blast/).

All primers were purchased from Sigma (UK) and were stored at -20°C.

43

Table 2.4. The forward and reverse primer sequences of all target genes used in the RT-qPCR

reaction.

Using MxPro software (Agilent Technologies, USA), the cycle threshold (Ct) was

determined in each sample, which is the number of cycles required for the fluorescent

signal to exceed the background level. Ct values are inversely proportional to the

amount of target nucleic acid in the sample. The 2-∆CT relative quantitation method

[218] was used to analyse the RT-qPCR data. Thus, in each sample, the expression of

the gene of interest is shown relative to β-actin (×1000).

Gene Accession

number Forward primer Reverse primer

β-actin NM_001101.3 ATGTACCCTGGCATTGCCGACA GACTCGTCATACTCCTGCTTGT

YAP1 NM_006106 TCCCGGGATGTCTGAGGAAT GGTTCGAGGGACACTGTAGC

P2RY6 NM_176798 GCCACCCACTATATGCCCTA GAAAAGGCAGGAAGCTGATG

MGMT NM_002412 TGGAGCTGTCTGGTTGTGAG CTGGTGAACGACTCTTGCTG

SLCO1B3 NM_019844 GGGTGAATGCCCAAGAGATA ATTGACTGGAAACCCATTGC

SLC36A4 NM_152313 CTGCCACCTTTGGTTGAAAT CTGTGGAGTGCCAGCTACAA

ZNF600 NM_198457 AACAGGGCAAGGCAATACAG GTGCTTCATGGCCATTTCTT

BIRC2 NM_001166 CACCATCAGAATTGGCAAGA ATTCGAGCTGCATGTGTCTG

LARP1B NM_178043 TTGCCTATTTCCCTGATTGC GGCCTGGTACAAACTCTGGA

44

2.13. siRNA-MEDIATED GENE KNOCK-DOWN IN THE PJ41 CELL LINE

2.13.1. KDalert™ GAPDH Assay kit for detection of GAPDH knock-down

The KDalert™ GAPDH Assay kit (Life Technologies, UK) includes the reagents

needed to detect silencing of glyceraldehyde-3-phosphate dehydrogenase (GAPDH)

in cultured cells at the protein level. The assay serves as a marker for cellular toxicity

resulting from transfection, and enables the optimal assay conditions to be identified.

The PJ41 SCCHN cell line was re-suspended in cell culture medium at two different

seeding densities (Table 2.5), and stored at 37°C until ready to use.

Table 2.5. Cell seeding densities used

Cell seeding density per mL Cell seeding density per well (80µL/well)

6.4×105 8.0×103

9.6×105 1.2×104

The siPORT Neo FX transfection agent (Life Technologies, UK) was diluted in Opti-

MEM medium (Life Technologies, UK) to a total volume of 10µL/well at three

different concentrations, as shown in Table 2.6. The samples were mixed and

incubated for 10 minutes at room temperature.

Table 2.6: Volumes of siPORT NeoFX and Opti-MEM used per well

GAPDH siRNA and negative control#1 siRNA (Life Technologies, UK) were re-

suspended to 2µM in nuclease-free water (Life Technologies, UK). 1.5µL siRNA

was then mixed with 8.5µL Opti-MEM (Life Technologies, UK) to give a total

volume of 10µL/well, and incubated for 10 minutes at room temperature.

The siPORT NeoFX samples were mixed with each siRNA in a 1:1 volume ratio and

incubated for 10 minutes at room temperature. 20µL of the transfection-complex was

dispensed into a 96-well tissue culture plate in triplicate. As a non-transfected

control, 20µL of Opti-MEM (Life Technologies, UK) was added to separate wells in

triplicate. 80µL of each cell suspension was dispensed into the wells and the plate

was incubated at 37°C for 24 hours, before replacing with 100µL fresh cell culture

siPORT NeoFX (µL/well) Opti-MEM (µL/well)

0.2 9.8

0.5 9.5

0.8 9.2

45

medium for a further 24 hours. Cells were checked under a light microscope for

cytotoxicity. A KDalert™ master mix was then prepared on ice (Table 2.7).

Table 2.7. Volumes of each component of the KDalert™ master mix used per well

Component Volume (µL/well)

KDalert™ Solution A 88.80

KDalert™ Solution B 0.68

KDalert™ Solution C 0.47

The 96-well plate containing the transfected cells was aspirated, replaced with

100µL/well lysis buffer and incubated at 4°C for 20 minutes. The lysates were

homogenised by pipetting up and down, before transferring 10µL to a new 96-well

plate. 10µL of water was also added to separate wells. 90µL of the master mix was

added to each well, mixed, and incubated for 15 minutes at room temperature. The

absorbance was measured using the Variskan® Flash plate reader (Thermo Scientific,

UK) at wavelength 620nm. The average absorbance for each sample (A620-sample) was

subtracted from the average absorbance of the water + master mix control (A620-WMM)

to determine a GAPDH activity (ΔA620-sample):

ΔA620-sample = A620-WMM − A620-sample

Next, the % remaining expression of GAPDH in each sample was calculated by

dividing the GAPDH activity in GAPDH siRNA-transfected cells (ΔA620-GAPDH) by

the GAPDH activity in negative control#1 siRNA-transfected cells (ΔA620-Negative):

% remaining expression = 100 x (ΔA620-GAPDH ÷ ΔA620-Negative)

The % GAPDH knock-down was then calculated:

% GAPDH knock-down = 100 − % remaining expression

The Optimal transfection conditions were those that maximised GAPDH knock-down

whilst lessening transfection-associated toxicity. This was predicted by using the

Optimal Balance Factor (OBF) (KDalert™ GAPDH assay kit User Guide, Thermo

Fisher website: https://tools.thermofisher.com/content/sfs/manuals/1639M.pdf). The

GAPDH activity in the negative control#1 siRNA-transfected cells (ΔA620-Negative) was

multiplied by the % GAPDH knock-down. The optimal conditions were those that

showed the greatest OBF value:

OBF = ΔA620-Negative × % GAPDH knock-down

46

2.13.2. siRNA-mediated knock-down of a target gene

Like the KDalert™ GAPDH assay, this procedure uses RNA interference (RNAi), a

biological mechanism found in eukaryotic cells. During this pathway, long dsRNA is

cleaved by the cytoplasmic nuclease Dicer, into small interfering RNAs (siRNAs).

The siRNA unwinds into two single-stranded RNAs (ssRNAs). The passenger strand

is degraded, whereas the guide strand assembles onto RNA-induced silencing

complexes (RISCs) that then pairs with a complementary messenger RNA (mRNA)

molecule. The catalytic component of the RISC complex cleaves the mRNA, leading

to specific gene silencing.

In this assay, standard cell culture media without FBS and antibiotics was used, which

will be referred to as media*. PJ41 cells were re-suspended to 9.6×105 cells/mL

(1.2×104 cells/well) in cell culture medium and stored at 37°C. 0.5µL/well siPORT

Neo FX (Life Technologies, UK) was mixed with 9.5µL/well media* and incubated

for 10 minutes at room temperature. 1.5µL/well 2µM negative control#1 siRNA (Life

Technologies, UK) or 2µM siRNA for the gene of interest (Table 2.8) were mixed

with 8.5µL/well media* and incubated for 10 minutes at room temperature. The

negative control#1 siRNA contained a nonsense sequence that served to monitor any

non-specific effects caused by unintended off-targeting.

Table 2.8. The siRNAs used in this study. Two different siRNAs were used for each target gene.

Target gene siRNA ID Supplier

SLCO1B3 s26261 Life Technologies, UK

s26262 Life Technologies, UK

MGMT s8750 Life Technologies, UK


SLC36A4 s42350 Life Technologies, UK


YAP1 s20366 Life Technologies, UK


ZNF600 s46376 Life Technologies, UK


P2RY6 sc-42584 Santa Cruz Biotechnology, USA


BIRC2 s1448 Life Technologies, UK


LARP1B s30246 Life Technologies, UK


Next, siPORT Neo FX was mixed with an equal volume of each siRNA or media* for

10 minutes at room temperature. The sample containing transfection agent and

47

media* served as an endogenous positive control for the protein of interest and as a

negative control for siRNA silencing, and enabled the cytotoxicity effects caused by

the transfection agent to be determined. 80µL of cell suspension and 20µL of each

transfection complex was added to each well of the plate, giving a total volume of

100µL/well. 20µL media* alone was also added to separate wells containing 80µL

cells, which served as an additional positive control for the protein of interest and a

negative control for siRNA silencing. The plate was incubated at 37°C for 24 hours,

before replacing the wells with 100µL fresh cell culture medium for a further 24

hours. Cells were then viewed under a light microscope for cytotoxicity.

Subsequently, RNA was extracted from the cells (Section 2.10) for cDNA synthesis

(Section 2.11) and RT-qPCR (Section 2.12) to determine the mRNA expression of

each target gene. Alternatively, cells were lysed for western blot analysis (Section

2.14) to evaluate YAP1 knock-down efficiency at the protein level.

2.13.3. siRNA-mediated knock-down of a target gene and infection with reovirus

The procedure described in Section 2.13.2 was followed to knock-down the gene of

interest by siRNA-mediated transfection. 24 hours after the final media change, the

cells were checked under a light microscope for cytotoxicity. Ten wells from each

treatment condition were carefully trypsinised, spun and re-suspended in working

media, before counting the cells using a haemocytometer. The cell counts were used

to calculate the volume of reovirus required for a certain multiplicity of infection

(MOI) in each well (Section 2.8), and then subsequent serial dilutions were made in

working media. The wells of the plate were replaced with 100µL diluted reovirus, or

100µL working media (un-infected sample) in triplicate, and incubated at 37°C for the

appropriate time-period. The development procedure described in Section 2.9 using

the Cell Titre 96® AQueous One Solution Reagent (Promega, UK) was then followed.

The average OD of the background control was subtracted from the average OD of

each sample. The % cell survival in each treatment condition was calculated by

dividing the reovirus infected sample by the un-infected sample, multiplied by 100:

% cell survival = (average OD infected sample ÷ average OD un-infected sample) × 100

48

2.14. WESTERN BLOTTING FOR PROTEIN DETECTION IN CELL

LYSATES

2.14.1. Lysate preparation and protein separation by sodium dodecyl sulfate

polyacrylamide gel electrophoresis (SDS-PAGE)

To prepare cell lysates, cells were washed twice with cold PBS (×1) (Fisher Scientific,

UK) and then lysed using RIPA buffer (Life Technologies, UK), which sometimes

contained the protease and phosphatase inhibitor cocktail mix (Life Technologies,

UK), diluted 1:100. After 5 minutes of gentle shaking on ice, the cell lysates were

collected and sheared three times with 21 gauge needles and 1mL syringes. Lysates

were then centrifuged at 13,000rpm for 5 minutes, before transferring the supernatants

to clean tubes and storing at -80°C. Total protein in the lysates was determined using

the Pierce™ BCA Protein Assay kit (Life Technologies, UK) by following the

manufacturer’s protocol. In brief, 25µL samples or albumin-containing standards

were added in duplicate to a 96-well plate, before adding 200µL working reagent for

30 minutes at 37°C. The purple-coloured reaction product was measured at 562nm on

the Variskan® Flash plate reader (Thermo Scientific, UK), and the protein

concentration in each sample was interpolated from the standard curve.

The samples were then diluted in RIPA buffer (Life Technologies, UK) to the

required concentration to ensure uniform protein loading. 13µL of the diluted lysate,

2µL NuPAGE® sample reducing agent (Life Technologies, UK) and 5µL NuPAGE®

LDS loading buffer (Life Technologies, UK) was mixed and placed on a heating

block at 70°C for 10 minutes. The XCell Surelock™ Mini-Cell Electrophoresis

apparatus (Life Technologies, UK) was assembled, containing NuPAGE® 4-12% Bis-

Tris gels (Life Technologies, UK) and NuPAGE® MOPS or MES SDS running

buffer (Life Technologies, UK), diluted 1:20 in water. 10µL Novex® Sharp pre-

stained markers (Life Technologies, UK) and 20µL lysate samples were loaded onto

the lanes of the gels. Electrophoresis was carried out at 200V for 1 hour using a

PowerPac (Bio-Rad, UK).

2.14.2. Protein transfer, blocking, antibody probing and band detection

Proteins from the gel were transferred to a nitrocellulose membrane by electroblotting

at 20 volts for 7 minutes using the iBlot® gel transfer device (program 3) (Life

49

Technologies, UK). The membrane blot was incubated overnight at 4°C in blocking

buffer (PBS/0.1% Tween-20 (Sigma, UK) containing 5% milk powder), with gentle

shaking. It was then probed with the required primary antibody diluted in blocking

buffer for 2 hours at room temperature. After washing for 3 × 10 minutes with

PBS/0.1% Tween-20, the appropriate horseradish peroxidase (HRP)-conjugated

secondary antibody (diluted in blocking buffer) was added to the blot for 2 hours at

room temperature. The optimum working antibody dilution was established prior to

use (Table 2.9). This was followed by another 3 × 10 minute wash in PBS/0.1%

Tween-20. The blot was then covered in the SuperSignal® West Pico

chemiluminescent substrate (Life Technologies, UK) for 3 minutes, before being

exposed to Carestream® Kodac® BioMax® autoradiography film (Sigma, UK) for an

appropriate period of time. The film was then dipped into Fixer and Developer

reagent (Sigma, UK) to observe protein bands. Alternatively, after exposure to

chemiluminescent substrate, protein bands were visualised using the ChemiDoc-It2

imager (UVP, UK). The molecular weight of the protein bands was confirmed by

direct comparison to the Novex® Sharp pre-stained marker. The blot was placed back

into PBS/0.1% Tween-20 and stored at 4°C until ready for stripping.

2.14.3. Stripping the membrane and antibody re-probing

To ensure uniform protein loading in each lane, the blot was stripped and re-probed

with β-actin or α-tubulin primary antibodies (Sigma, UK), whose expression should

remain constant in the test samples. 15mL of Restore™ PLUS stripping buffer

(Fisher Scientific, UK) was added to the blot with gentle agitation for 15 minutes,

before washing for 3 × 5 minutes with PBS/0.1% Tween. The blot was then blocked,

re-probed with the relevant antibodies, and developed as described in Section 2.14.2.

2.14.4. Densitometry analysis

The scanned images of the blots were uploaded to the Image Studio™ Lite Version

4.0 software (LI-COR Biotechnology, UK), and the signal for each protein band was

quantified. To directly compare the protein expression level between samples, the

band density of the protein of interest was normalised to the band density of the

loading control in their respective lanes. The relative protein density in each sample

was then plotted graphically.

50

Table 2.9. Primary and secondary antibodies used for western blotting.

2.15. BACTERIAL TRANSFORMATION AND PURIFICATION OF

PLASMID DNA

Plasmid DNA vectors were used to deliver and over-express the YAP1 gene in cell

lines. To generate enough plasmid DNA, bacterial transformation was performed.

Bacteria are used as hosts for making copies of DNA, as they are easy to grow and

their cellular machinery naturally carries out DNA replication and protein synthesis.

2.15.1. Bacterial transformation and making a glycerol stock

Bacterial growth mediums, LB Broth with agar (Sigma, UK) and LB Broth (Sigma,

UK), were dissolved in water (1 tablet in 50mL water) and autoclaved. After

warming the LB Broth with agar to 50°C in a waterbath, Ampicillin (Sigma, UK) was

added at a dilution of 1:1000 (for example, 50µL ampicillin in 50mL LB agar), under

sterile conditions. 25mL was added to non-tissue culture petri dishes. Once set, the

petri dishes were pre-warmed to 37°C, and the DNA plasmid was diluted in sterile,

autoclaved water to 1ng/µL. The One Shot® TOP10 chemically competent E.Coli

Primary antibodies

Antibody Supplier Dilution Molecular

Weight (kDa)

Anti-YAP1 mouse monoclonal IgG1 Abcam, UK 1:2000 54

Anti-phospho-YAP (S127) rabbit polyclonal IgG Cell Signalling, USA 1:1000 65

Anti-T3D rabbit polyclonal IgG reovirus antiserum

Cocalico Biologicals, USA.

Kindly donated by Dr Dermody,

Vanderbilt University, USA.

1:2000

λ = 160

µ = 80

σ = 40

Anti-acetylated α-tubulin mouse monoclonal IgG2b Sigma, UK 1:30000 50

Anti-β-actin mouse monoclonal IgG1 Sigma, UK 2000 42

Anti-α-tubulin mouse monoclonal IgG1 Sigma, UK 1:4000 50

Secondary antibodies

Antibody Supplier Dilution -

Rabbit anti-mouse IgG-HRP Life Technologies, UK 1:2000 -

Goat anti-rabbit IgG-HRP Cell Signalling, USA 1:2000 -

51

protocol (Life Technologies, UK) was then followed, according to the manufacturer’s

instructions. In summary, one vial of competent cells was thawed on ice, before

adding 1-5µL DNA. The vial was mixed gently by inversion and incubated on ice for

30 minutes. The cells were heat-shocked in a 42°C waterbath for 30 seconds and then

placed directly onto ice for 2 minutes. 250µL of S.O.C medium was added aseptically

to the vial and incubated in the orbital shaker at 37°C for 1 hour at 225rpm.

Subsequently, 5µL, 20µL or 200µL of each transformation was spread onto the pre-

warmed petri dishes using a cell scraper, to ensure that at least one dish would have

well-spaced colonies. A negative control dish was also included that did not contain

any transformation. The dishes were incubated for 24 hours at 37°C. Colonies were

then picked with a sterile pipette tip and placed into a T-150 tissue culture flask

containing 120mL LB Broth (Sigma, UK) and 120µL Ampicillin (Sigma, UK). The

flask was incubated in the orbital shaker at 37°C and 100rpm for 24 hours. A

negative control flask containing LB Broth and Ampicillin (with no colony) was

included to check for contamination. To make a bacterial glycerol stock for long-term

storage of the plasmid, 500µL of the liquid bacterial culture was added to a 1.5mL

Eppendorf tube containing 500µL of 100% glycerol (Sigma, UK). The tube was

mixed and frozen at -80°C.

2.15.2. Purification of the plasmid DNA

The plasmid DNA was isolated from the bacterial culture using the QIAfilter Plasmid

Midi prep kit (Qiagen, UK), following the manufacturer’s instructions. In brief, the

bacterial cells in the liquid culture were pelleted by centrifugation and subsequently

lysed using a series of buffers. Bacterial lysates were incubated in a QIAfilter

cartridge and cleared by filtration. The cleared lysate was then loaded onto an anion-

exchange QIAGEN-tip where the plasmid DNA could selectively bind, under

appropriate low-salt and pH conditions. Impurities were removed by a medium-salt

wash and then the plasmid DNA was eluted in a high-salt buffer. Consequently, the

DNA was concentrated and de-salted by isopropanol precipitation. After centrifuging

the DNA sample, the resulting pellet was collected and air-dried for 1 minute, before

dissolving it in 100µL sterile, autoclaved water. The concentration of the purified

plasmid DNA was determined using the NanoDrop® ND-1000 Spectrophotometer

(Labtech International, UK), and stored at -20°C.

52

2.16. LIPID-MEDIATED OVER-EXPRESSION OF YAP1 IN CELL LINES

2.16.1. Transient over-expression of YAP1

In order to over-express the YAP1 gene, cell lines were transfected with a DNA

plasmid containing a YAP1 cDNA insert. Lipofectamine® is a cationic lipid

transfection reagent that mediates the interaction of the nucleic acid with the

negatively charged cell surface membrane. Once bound to the cell membrane, the

liposome/nucleic acid complex enters the cell via endocytosis, which then diffuses

through the cytoplasm and into the nucleus for gene expression.

Cells were seeded (100µL/well) in 96-well plates to the appropriate seeding density in

cell culture media and incubated overnight at 37°C. The cell concentration was

optimised to ensure an 80% confluence on the day of the transfection. Plasmid DNA

was transformed and purified (Section 2.15), and then reconstituted in sterile,

autoclaved water to 100ng/µL. A DNA mix was prepared by adding plasmid DNA,

Opti-MEM (Life Technologies, UK) and PLUS reagent (Life Technologies, UK), as

shown in Table 2.10. A ‘no plasmid DNA’ control containing Opti-MEM (Life

Technologies, UK) and PLUS reagent (Life Technologies, UK) was also prepared.

The samples were mixed and incubated at room temperature for 5 minutes.

Table 2.10. The preparation of plasmid DNA and the ‘no plasmid DNA’ control.

Condition 100ng/µL Plasmid

DNA (µL/well) Opti-MEM (µL/well)

PLUS reagent

(µL/well)

Plasmid DNA + transfection 1 8.85 0.15

No plasmid DNA control - 9.85 0.15

The transfection conditions were optimised in the PJ34 SCCHN cell line by using

three different concentrations of Lipofectamine® LTX (Life Technologies, UK)

diluted in Opti-MEM (Life Technologies, UK), as displayed in Table 2.11. In the

COS-1 cell line, a single concentration of 0.45µL/well Lipofectamine® was used.

The samples were mixed and incubated at room temperature for 5 minutes.

Table 2.11. The preparation of Lipofectamine® LTX.

Lipofectamine® LTX (µL/well) Opti-MEM (µL/well)

0.25 9.75

0.35 9.65

0.45 9.55

53

Each Lipofectamine® sample was mixed with an equal volume of the plasmid DNA

or ‘no plasmid DNA’ samples, and incubated for 30 minutes at room temperature.

The ‘no plasmid DNA’ sample therefore served as a control for any cytotoxicity

caused by the transfection agent alone. Meanwhile, the wells of the 96-well plate

were aspirated and replaced with 100µL Opti-MEM (Life Technologies, UK), before

returning it to the 37°C incubator. The wells were then replaced with 20µL

transfection mix and 80µL Opti-MEM (Life Technologies, UK). 100µL DMEM cell

culture medium was also added to separate wells of the plate for use as a un-

transfected control. The contents of the wells were gently mixed on a plate shaker

and then incubated for 5.5 hours at 37°C. The transfection was aspirated from the

wells and replaced with 100µL cell culture medium before returning the plate to the

37°C incubator for 24 hours. The visual appearance of the cells was checked under a

light microscope for signs of cytotoxicity. To determine the mRNA expression of

YAP1, RNA was extracted from the cells (Section 2.10) for cDNA synthesis (Section

2.11) and RT-qPCR analysis (Section 2.12). To evaluate YAP1 over-expression at

the protein level, cells were lysed for western blot analysis (Section 2.14). For the

PJ34 cell line, the optimal concentration of Lipofectamine® was 0.45µL/well, which

ensured limited cytotoxicity whist maintaining a reasonable expression level of YAP1.

2.16.2. Stable over-expression of YAP1 in the HN5 SCCHN cell line

Unlike transient transfection where the introduced DNA persists in cells only for

several days, stable transfection introduces DNA into cells long-term. This is because

the transfected DNA has been incorporated into the cells’ genome and thus, is passed

into their progeny. To determine the optimal antibiotic concentration for selecting

stable cell colonies, a dose-response experiment (kill curve) was performed using the

G418 (Geneticin disulfate salt) antibiotic (Sigma, UK). Cells were seeded to the

required density in cell culture medium in a 24-well plate for 24 hours at 37°C. G418

was serially diluted in cell culture medium to make an 8-point dilution curve. The

wells of the plate were aspirated and replaced with 1mL diluted G418 in duplicate,

before returning to the 37°C incubator. A ‘no antibiotic’ control was also added to

separate wells in duplicate, which contained cell culture media only. The wells were

replaced with fresh G418-containing media every 2 days for up to one week, and the

cells were examined under a light microscope each day for signs of toxicity. The

54

optimal G418 concentration was the lowest concentration that killed all cells within 1

week. In the case of the HN5 cell line, this was 650µg/mL.

Having found the optimal concentration of G418, the protocol described in Section

2.16.1 was followed to transfect cells with the required DNA plasmid in 96-well

plates. 24 hours after the final media change, the cells were examined under a

microscope for signs of toxicity. 100µL trypsin was added to each well to detach

adherent cells from the surface. The cells were centrifuged at 1500rpm for 3 minutes

and re-suspended in 1mL cell culture medium, which was added to 10cm tissue

culture petri dishes containing 10mL cell culture media. The dishes were incubated

for 24 hours at 37°C, before replacing with fresh cell culture medium containing the

650µg/mL G418 antibiotic, and returning to the 37°C incubator. Fresh G418-

containing media was replaced every 2-3 days until circular colonies formed. Since

the DNA plasmids used for transfection contained the neomycin antibiotic resistance

gene, it was assumed that these surviving colonies contained the transfected gene of

interest, whereas cells that failed to uptake the plasmid were killed by G418. Under

sterile conditions, the colonies were picked with a pipette tip and transferred to wells

of a 96-well plate. The cells were transferred to culture wells of increasing surface

area containing G418-media until confluent in T-150 flasks, before being stored in

liquid nitrogen (Section 2.6). Several stable clones were lysed for western blot

analysis (Section 2.14) to determine YAP1 over-expression at the protein level,

compared to the parental HN5 cell line or empty-vector (EV) stable clones. All DNA

plasmid sequences used for transfection were validated by the Sanger sequencing

facility, University of Cambridge, UK, and are listed in Table 2.12.

Table 2.12. The DNA plasmids used for transient or stable transfection of cell lines.

DNA plasmid name cDNA

insert Vector type Supplier

Promoter

type

Selection

marker Reference

Human Flag-tagged YAP1 YAP1 pcDNA™3.1 A gift from Dr Nic Tapon,

Cancer Research UK. CMV Neomycin [219]

Human EYFP-tagged YAP1 YAP1 pBK-CMV

phagemid vector

A gift from Dr Nic Tapon,

Cancer Research UK. CMV Neomycin

Empty vector (EV) control - pcDNA™3.1 A gift from Dr Lisi Meira,

University of Surrey, UK. CMV Neomycin [219]

55

2.16.3. Over-expression of YAP1 and reovirus infection

Cell lines were transiently or stably transfected with the required DNA plasmid in 96-

well plates, as described in Sections 2.16.1 and 2.16.2 respectively. Ten wells

containing cells from each treatment condition were carefully trypsinised, spun, re-

suspended in 200µL working media, and counted using a haemocytometer. Cell

counts were used to calculate the volume of reovirus required for a certain MOI per

well (Section 2.8). From this, subsequent serial dilutions of reovirus were made in

working media. The plate was aspirated and replaced with 100µL/well diluted

reovirus or 100µL/well working media (un-infected sample) in triplicate, and

incubated at 37°C for the appropriate period of time. The development procedure

described in Section 2.9 with the Cell Titre 96® AQueous One Solution Reagent

(Promega, UK) was then followed. The average OD of the background control was

subtracted from the average OD of each sample. The % cell survival in each

treatment condition was calculated by dividing the reovirus infected sample by the un-

infected sample, multiplied by 100:


2.17. DETECTION OF A PROTEIN USING IMMUNOFLUORESCENCE

STAINING AND CONFOCAL MICROSCOPY IN CELL LINES

Immunofluorescence staining was used to demonstrate both the presence and cellular

localisation of total YAP1, phospho-YAP-S127, or reovirus proteins in SCCHN cell

lines. Cells were seeded (500µL/well) in 8-well tissue culture-treated glass chamber

slides (BD BioSciences, UK). The chamber slides were incubated for 24 hours at

37°C to ensure an 80% confluence. Wheat Germ Agglutinin (WGA) Alexa Fluor®

594 conjugate (Life Technologies, UK), a plasma membrane stain, was diluted 1:200

in warm Hanks media (Sigma, UK). The wells of the chamber slide were aspirated

replaced with 300µL/well of WGA for 10 minutes at 37°C. To fix the cells, the WGA

stain was removed and replaced with 300µL warm 4% paraformaldehyde for 10

minutes at 37°C (made fresh in-house by adding 20mL PBS (×1) (Fisher Scientific,

UK) and 0.8g paraformaldehyde (Sigma, UK) on a magnetic stirrer at 60°C until

dissolved). The wells were then rinsed three times with 500µL PBS (×1) (Fisher

56

Scientific, UK), before permeabilising the cells with 300µL 0.2% Triton X-100

(Sigma, UK) for 10 minutes at 37°C.

After three rinses in 500µL PBS (×1) (Fisher Scientific, UK), non-specific binding

sites were blocked in 300µL/well 10% Goat serum diluted in PBS (×1) (Fisher

Scientific, UK) for 20 minutes at 37°C. The wells were aspirated and 300µL/well of

the required primary antibody, diluted in PBS/1%BSA, was added for 2 hours at

37°C. A ‘no primary antibody’ sample was also included containing 300µL

PBS/1%BSA for use as a negative control. The wells were rinsed three times with

500µL PBS (×1) (Fisher Scientific, UK), before simultaneously adding 300µL/well of

the appropriate Alexa Fluor® secondary antibody and the TOPRO®-3 nuclear stain

(Life Technologies, UK), diluted 1:400 in PBS/1%BSA, for 1 hour at 37°C. Both the

WGA and TOPRO®-3 stains helped to localise the protein of interest in the cells.

The wells were then aspirated and rinsed three times with 500µL PBS (×1) (Fisher

Scientific, UK). The plastic chamber was removed carefully using the white comb

supplied (BD BioSciences, UK), revealing the glass slide underneath. 2 drops of

Vectashield for fluorescence (Vector Laboratories, UK) was added to the slide before

being cover-slipped. The cells were imaged after 24 hours using the Nikon A1M

confocal microscope and NIS elements acquisition software (Nikon, UK). Primary

and secondary antibodies used in this procedure are listed in Table 2.13.

Table 2.13. Primary and secondary antibodies used for immunofluorescence staining.

Primary antibodies

Antibody Supplier Dilution

Anti-YAP1 mouse monoclonal IgG1 Abcam, UK 1:200

Anti-phospho-YAP (S127) rabbit polyclonal IgG Cell Signalling, USA 1:200


Cocalico Biologicals, USA. Kindly

donated by Dr Dermody, Vanderbilt

University, USA.

1:1000



Alexa Fluor® 488 goat anti-mouse IgG1 Life Technologies,UK 1:200

Alexa Fluor® 488 goat anti-rabbit IgG Life Technologies,UK 1:200


57

2.18. THE EFFECT OF SPHINGOSINE-1-PHOSPHATE (S1P) ON YAP1

ACTIVITY AND REOVIRUS ONCOLYSIS

S1P has been previously shown to cause de-phosphorylation of the YAP1 protein on

residue serine 127 (S127). This consequently resulted in the nuclear migration of

YAP1 in certain cell lines [220]. To determine the effect of S1P treatment on reovirus

oncolysis, the PJ41 cell line was re-suspended to a concentration of 1×105cells/mL.

100µL/well cell suspension was added to a 96-well plate for 24 hours at 37°C to

ensure an 80% confluence. Cells were then treated with 100µL/well S1P (Sigma,

UK) diluted to 1µM in working cell culture media, or with media alone (un-treated

cells), for 60 minutes at 37°C. After this time, ten wells containing cells from each

treatment condition were carefully trypsinised, spun, re-suspended in 200µL working

media, and counted using a haemocytometer. Cell counts were used to calculate the

volume of reovirus required for MOI 500 per well (Section 2.8). From this,

subsequent serial dilutions of reovirus were made in working media. The plate was

aspirated and replaced with 100µL/well of each dilution of reovirus, or with

100µL/well working media (un-infected cells) in triplicate, and incubated at 37°C for

24 hours. The development procedure described in Section 2.9 with the Cell Titre

96® AQueous One Solution Reagent (Promega, UK) was then followed. The average

OD of the background control was subtracted from the average OD of each sample.

The % cell survival in each treatment condition was calculated by dividing the

reovirus infected sample by the un-infected sample, multiplied by 100:


In order to assess the de-phosphorylation of YAP1 caused by S1P, PJ41 cells were

seeded in 96-well plates at 1×105cells/mL (100µL/well) for 24 hours at 37°C. Cells

were then treated with 100µL/well S1P (Sigma, UK) diluted to 1µM for 20, 30 and 60

minutes at 37°C, or with media alone for 60 minutes as an un-treated control. Cells

were then lysed for western blot analysis (Section 2.14) to determine the protein

expression levels of phosphor-YAP-S127 and total YAP1 in each sample.

58

2.19. DETECTION OF A PROTEIN BY INDIRECT FLOW CYTOMETRY

Flow cytometry was used to analyse the expression of cell surface JAM-A, as well as

intracellular YAP1 and reovirus proteins in SCCHN cell lines. Cells were harvested

and re-suspended to a concentration of 2×106cells/mL in ice-cold FACS buffer

(prepared in-house with PBS (×1) (Fisher Scientific, UK), 10% bovine serum albumin

(BSA) (Sigma, UK) and 1% sodium azide (Sigma, UK)). 200µL cell suspension was

added to FACS tubes. The tubes were centrifuged at 2000rpm for 2 minutes and the

supernatant was poured off. The cell pellet was re-suspended in 200µL/tube 80%

methanol (Fisher Scientific, UK) and incubated for 5 minutes at room temperature.

This enabled cells to be fixed, allowing the target protein to be retained in the original

cellular location. The methanol fixative was washed off by adding 1mL ice cold

PBS/0.1% Tween to each tube that were then centrifuged at 2000rpm for 2 minutes.

After discarding the supernatant, the cells were permeabilised with 200µL/tube

PBS/0.1% Tween and incubated for 20 minutes at room temperature. The tubes were

centrifuged at 2000rpm for 2 minutes and the supernatant was removed. Non-specific

binding sites were blocked by adding 100µL/tube 10% normal goat serum (Dako,

UK) diluted in PBS (×1) (Fisher Scientific, UK) and incubated for 10 minutes at room

temperature. 1mL ice cold PBS/0.1% Tween was added to each tube and

subsequently centrifuged at 2000rpm for 2 minutes. After discarding the supernatant,

the primary antibody was diluted in FACS buffer to the appropriate concentration.

50µL was added to each tube and incubated for 30 minutes at room temperature.

Negative control samples were treated with 50µL FACS buffer alone without primary

antibody. 1mL ice cold PBS/0.1% Tween was added to each tube and centrifuged at

2000rpm for 2 minutes, before discarding the supernatant. 100µL/tube Alexa Fluor®

secondary antibody (Life Technologies, UK) diluted in FACS buffer, was added and

incubated in the dark for 30 minutes at room temperature. Following a final wash in

1mL ice cold PBS/0.1%, each tube was centrifuged at 2000rpm for 2 minutes and the

supernatant was poured off. Cells were re-suspended in 200µL FACS buffer before

being analysed by flow cytometry on the MACS Quant® Analyser (Miltenyi Biotec,

UK). The Mean Fluorescence Intensity (MFI) of the negative control sample was

subtracted from the MFI value of each positive sample, resulting in an overall MFI

protein expression value. Primary and secondary antibodies used in this procedure are

listed in Table 2.14.

59

Table 2.14. Primary and secondary antibodies used for indirect flow cytometry protein detection.

2.20. ONE STEP GROWTH CURVE ANALYSIS BY THE 50% TISSUE

CULTURE INFECTIVE DOSE (TCID50) ASSAY

TCID50 is a measure of infectious virus titre. In this endpoint dilution assay, the

amount of virus needed to kill 50% of infected host cells was quantified. TCID50 is

not equivalent to plaque forming unit (pfu) due to differences in assay methods

(Section 2.24), although the theoretical relationship is 0.69 pfu = 1 TCID50 based on

the Poisson distribution (Section 2.8).

2.20.1. Preparation of intracellular or extracellular viral samples

To prepare viral samples for the assay, the relevant SCCHN cell line was seeded in

96-well tissue culture plates for 24 hours at 37°C, to reach 80% confluency. The cells

were then infected with reovirus at MOI 5 (Section 2.8) for different time increments

of up to 72 hours. Intracellular viral samples were made by removing the supernatant

from the wells and adding 100µL/well fresh cell culture medium. Subsequently, the

plate was freeze-thawed three times to release cellular viral particles into the culture

medium. Extracellular viral samples were made by transferring the viral supernatant

to a 1.5mL Eppendorf tube, which was centrifuged at 300 × g for 2 minutes to remove

cell debris. The resulting supernatant was carefully removed and transferred to a fresh

1.5mL Eppendorf tube.

Primary antibodies


Anti-YAP1 mouse monoclonal IgG1 Abcam, UK 1:200

Anti-JAM-A rabbit polyclonal IgG Santa Cruz Biotechnology, USA 1:200


Cocalico Biologicals, USA. Kindly

donated by Dr Dermody, Vanderbilt

University, USA.

1:1000



Alexa Fluor® 546 goat anti-mouse IgG1 Life Technologies,UK 1:500


60

2.20.2. Infection of the host-cell monolayer and determining the cytopathic effect

Host L929 cells were seeded (100µL/well) in 96-well tissue culture plates at a density

of 2×105 cells/mL in cell culture medium for 24 hours at 37°C, to ensure a 90%

confluence. 1:10 dilutions of the relevant reovirus sample was made in working

media. For example, using a sterile filtered tip, 150µL of the viral sample was mixed

with 1350µL working media to make a 1×10-1 dilution. With a new tip, 150µL of this

mixture was transferred to the next tube containing 1350µL working media to make

1×10-2 dilution. The series was repeated through to a 1×10-9 dilution. The 96-well

plates were aspirated and infected with 100µL of each virus sample (12 wells per

dilution), or with working media for use as a non-infected control. The virus was

absorbed for 3 hours before replacing each well with 100µL cell culture medium for 3

days at 37°C. Using a light microscope, the cytopathic effect (CPE) in each well was

observed. The number of positive and negative wells was recorded for each dilution

and used to calculate the reovirus titre (TCID50/mL) in each sample, according to the

Spearman and Karber algorithm [221].

2.21. VERIKINE™ HUMAN INTERFERON BETA (IFN-β) ENZYME-

LINKED IMMUNOSORBENT ASSAY (ELISA)

IFN-β is secreted by fibroblasts and many other cell types in response to pathogens,

including viruses. Secretion of IFN-β is known to inhibit viral replication as part of

the body’s innate anti-viral response. The concentration of IFN-β in reovirus-infected

or non-infected cell line supernatants was quantified by using The Verikine™ Human

IFN-β ELISA kit (PBL Assay Science, USA), according to the manufacturer’s

instructions.

2.21.1. Isolation of peripheral blood mononuclear cells (PBMCs) from whole

blood

PBMCs (lymphocytes, macrophages and monocytes) were used as a positive control

for IFN-β secretion. In order to isolate PBMCs, 10mL whole blood from a healthy

human donor was diluted in 20mL Hanks’ balanced salt solution (Sigma, UK). This

was carefully layered onto 15mL Histopaque®-1077 (Sigma, UK) in a 50mL falcon

61

tube, which was then subjected to density-gradient centrifugation at 690 x g for 25

minutes. The PBMC layer was removed and placed into a 50mL Falcon tube

containing 30mL Hanks’ balanced salt solution (Sigma, UK) and centrifuged at 690g

for 10 minutes. The supernatant was discarded and the cell pellet was gently re-

suspended in 1mL Hanks’ balanced salt solution (Sigma, UK), before adding a further

29mL to the tube. After being centrifuged at 690g for 10 minutes, the supernatant

was removed and the pellet was re-suspended in 1mL RPMI cell culture medium.

The cells were counted with a haemocytometer, as described in Section 2.5.

2.21.2. Preparation of cell supernatants

100µL/well human SCCHN cell lines or PBMCs were seeded in 96-well tissue culture

plates and incubated for 24 hours at 37°C, to ensure an 80% confluence. The wells

were aspirated and replaced with 100µL working media as a non-infected control, or

with 100µL reovirus diluted to the desired MOI (Section 2.8) in working media for 24

hours at 37°C. The supernatants were then harvested in 1.5mL Eppendorf tubes and

centrifuged for 2 minutes at 300 × g to remove cell debris. The supernatants were

transferred to fresh tubes and stored at -80°C.

To determine whether over-expression or knock-down of YAP1 affected IFN-β

production, supernatants were also collected after stable cell line generation from the

parental HN5 cell line, or from siRNA-transfected PJ41 cells, infected with or without

reovirus (Sections 2.16.3 and 2.13.3 respectively).

2.21.3. The Verikine™ Human IFN-β ELISA assay

A 7-point human IFN-β standard curve was prepared, ranging from 4000 to 50pg/mL.

A blank sample was also included containing sample diluent alone, which served as a

background control. At room temperature, 50µL sample diluent was added to all

wells of the pre-coated IFN-β 96-well plate, before adding 50µL of the diluted

standards, blank or test samples for 1 hour, in duplicate. After three washes with the

provided buffer, 100µL of antibody solution was added to each well for 1 hour. The

plate was washed three times and then 100µL/well of HRP solution was added for 1

hour. Following a final three washes, 100µL/well of TMB substrate was added for 15

minutes in the dark. 100µL stop solution was dispensed into each well before reading

the absorbance at 450nm on the Variskan® Flash plate reader (Thermo Scientific,

62

UK), to generate OD raw data values. The average OD of the blank was subtracted

from the average OD of the standards and test samples. The OD values for the

standard curve were then plotted using a 4-parameter fit (GraphPad Prism version 6

software, USA), enabling the IFN-β titre in the test samples to be determined.

2.22. DETECTION OF THE YAP1 PROTEIN IN TISSUE BY ENZYMATIC

IMMUNOHISTOCHEMISTRY (IHC) STAINING

The presence and location of the YAP1 protein was visualised in intact head and neck

cancer or normal tissue sections by enzymatic IHC staining. The following tissue

microarrays were purchased from US Biomax (USA) and used for this purpose:

HN803a: a tissue microarray containing 60 cases of squamous cell carcinoma of

the head and neck, 1 case of head and neck sarcomatodes and 8 cases of head and

neck metastatic carcinoma. Additionally, the array contained 1 cancer adjacent

normal tissue and 10 normal tissues derived from the tongue.

FDA999c: a multiple organ normal tissue microarray containing 99 cases,

including 12 normal head and neck tissues.

Human PCa tissue was used as a positive control for the YAP1 protein, as

recommended by the manufacturer of the YAP1 primary antibody (Abcam, UK) and

published data [222]. PCa tissues were fixed in 10% neutral buffered formalin for 18-

24 hours, before being embedded in paraffin and cut on a microtome. Once affixed to

the slide, the tissue was dried at room temperature and then baked for 1 hour at 60°C.

The PCa tissue sample was obtained with patient consent from The Royal Surrey

Hospital. Tissue microarrays were not coated with an extra layer of paraffin, and

were therefore placed directly into xylene for the de-paraffinization procedure.

2.22.1. Deparaffinization and antigen retrieval

The removal of paraffin from the slides was imperative in order to achieve the desired

staining on the section. Slides were placed into 3 × 5 minute changes of 100% xylene

(Sigma, UK) followed by two changes in 100% ethanol (Fisher Scientific, UK),

before being placed in methanol (Sigma, UK) containing 0.3% hydrogen peroxide

63

(Sigma, UK) for 20 minutes to block endogenous peroxidases. The slides were

rehydrated in 100%, 70%, and 50% ethanol (Fisher Scientific, UK) and then placed

into distilled water. For the antigen retrieval step, the slides were subjected to boiling

in 0.01M citrate buffer (pH 6.0) and microwaved for 12 minutes on the high setting.

After this time, the slides were left to cool at room temperature in the citrate buffer for

1-2 hours. The antigen retrieval step serves to break the methylene bridges that are

formed during fixation, and helps to expose antigenic sites to allow antibodies to bind.

The slides were then washed in distilled water for 3 minutes, followed by 2 × 3

minute washes in PBS (×1) (Fisher Scientific, UK).

2.22.2. Blocking of the tissue and addition of antibodies

After placing the slides into a moist chamber, an ImmEdge pen (Vector Laboratories,

UK) was used to create a barrier around the tissue sections. The slides were blocked

by adding 3 drops of normal horse serum, as supplied in the RTU Vectastain

Universal Elite ABC kit (Vector Laboratories, UK), for 15 minutes. A further

blocking step was performed using the Avidin/Biotin blocking kit (Vector

Laboratories, UK). 3 drops per section of Avidin D solution was added for 15

minutes. Following a quick wash in PBS (×1) (Fisher Scientific, UK), the slides were

then exposed to 3 drops of biotin solution for 15 minutes. Each section was

subsequently incubated with 200µL anti-YAP1 mouse monoclonal IgG1 primary

antibody (Abcam, UK), diluted 1:400 in PBS/0.1% BSA, for 24 hours. As a negative

control, 200µL PBS/0.1% BSA alone was added to corresponding slides. The slides

were washed for 3 × 3 minutes in PBS (×1) (Fisher Scientific, UK), before adding 3

drops of universal secondary antibody, as supplied in the RTU Vectastain Universal

Elite ABC kit (Vector Laboratories, UK), to each section for 30 minutes. The

secondary antibody used in this kit was a cocktail of biotinylated anti-mouse IgG and

anti-rabbit IgG, designed for use with both rabbit and mouse primary antibodies.

2.22.3. Addition of the Avidin-biotin complex (ABC) and 3,3’diaminobenzidine

(DAB) substrate

Following a 3 × 3 minute wash in PBS (×1) (Fisher Scientific, UK), the slides were

placed back into the chamber. 3 drops of ABC reagent, supplied in the RTU

Vectastain Universal Elite ABC kit (Vector Laboratories, UK), was dropped onto the

64

sections for 30 minutes. DAB substrate solution was prepared by adding 4 drops of

DAB Peroxidase substrate solution (Vector Laboratories, UK) to 5mL distilled water.

After another 3 × 3 minute wash in PBS (×1) (Fisher Scientific, UK), 3 drops of the

diluted DAB solution was added to the sections for 2-10 minutes, which produced a

brown reaction product in the presence of HRP enzyme. The slides were then washed

in distilled water for 5 minutes. The sections were counterstained with 3 drops of

haematoxylin (Vector Laboratories, UK) for 45 seconds, before immediately washing

the slides with water under a running tap for 5 minutes. Haematoxylin is a blue

nuclear stain that provided contrast to the tissue.

2.22.4. Dehydration of the tissue section, cover-slipping and scoring

Slides were dehydrated in 50% ethanol, 70% ethanol, three changes of 100% ethanol

(Fisher Scientific, UK) and three changes of 100% xylene (Sigma, UK). After being

left to dry for 1 hour, 1 drop of Vector mounting media (Vector Laboratories, UK)

was added to each section. A glass coverslip (VWR International) was placed directly

over the top. The slides were placed on a hot rack for 30 minutes and stored at room

temperature, before being analysed under the light microscope.

A system was used to score the tissues according to the intensity of YAP1 brown

colouration, where 0 = negative, +1 = weak positive, +2 = moderate positive and +3 =

strong positive staining. The diagnosis, staining intensity, and localisation of YAP1

in all tissues was confirmed by a consultant head and neck pathologist at The Royal

Surrey Hospital (Dr Silvana Di Palma).

65

2.23. ASSESSING THE INTERACTION BETWEEN REOVIRUS AND

TAXANE CHEMOTHERAPY DRUGS IN PCa CELL LINES

The procedure described in Section 2.9 was followed in order to determine the IC50

values of reovirus, Cabazitaxel and Docetaxel in prostate cell lines. These values

were then used to establish the doses of each agent in combination assays.

2.23.1. Concurrent combination of two agents at fixed-dose ratios

DU145 and LNCaP PCa cell lines were seeded (100µL per well) in 96-well plates in

cell culture media for 24 hours at 37°C, to ensure an 80% confluence. Cells were

treated with 100µL/well working cell culture media as an un-treated control, or with

reovirus, Cabazitaxel or Docetaxel as single agents, at doses representing 4, 2, 1, 0.5

and 0.25 times their respective IC50 values. Reovirus was also combined with

Cabazitaxel or Docetaxel at these fixed-dose ratios and added to the wells

concurrently (100µL/well). All treatments were added in triplicate wells and

incubated at 37°C for 96 hours, before developing the plates with the Cell Titre 96®

AQueous One Solution Reagent (Promega, UK) (Section 2.9). After calculating the %

cell survival relative to un-treated cells, the fraction of dead cells affected in each

treatment condition was evaluated, as described in Section 2.9, and inputted into

CalcuSyn software (BioSoft, UK). The level of interaction between reovirus in

combination with Cabazitaxel or Docetaxel was determined by the Chou and Talalay

equation (Section 2.27.5).

2.23.2. Comparing sequential and concurrent combinations at fixed-dose ratios

The DU145 PCa cell line was seeded (100µL per well) in 96-well plates in cell culture

media for 24 hours at 37°C, to ensure an 80% confluence. Cells were treated with

working cell culture medium alone as an un-treated control, or with reovirus or

Cabazitaxel as single agents at doses representing 1.00, 0.50, 0.25, 0.13 and 0.06

×IC50 values. In addition, reovirus was combined with Cabazitaxel at these doses,

only this time concurrent combination treatment was directly compared to five

different sequential combinations, in order to determine the most synergistic

sequencing strategy. Cells were treated in triplicate for a total of 96 hours at 37°C.

Concurrent treatment: reovirus and Cabazitaxel added simultaneously

Sequential treatment 1: 1 hour Cabazitaxel alone, followed by reovirus

66

Sequential treatment 2: 4 hours Cabazitaxel alone, followed by reovirus

Sequential treatment 3: 24 hours Cabazitaxel alone, followed by reovirus

Sequential treatment 4: 24 hours reovirus alone, followed by Cabazitaxel

Sequential treatment 5: 48 hours reovirus alone, followed by Cabazitaxel

The plates were developed with Cell Titre 96® AQueous One Solution Reagent

(Promega, UK) and the % cell survival was calculated relative to untreated cells

(Section 2.9). The fraction of dead cells affected in each treatment was evaluated

(Section 2.9) and inputted into CalcuSyn software (BioSoft, UK). The interaction

between reovirus and Cabazitaxel was determined by the Chou and Talalay equation

(Section 2.27.5).

2.23.3. Concurrent combination of two agents at non-fixed dose ratios

The DU145 PCa cell line was seeded (100µL per well) in 96-well plates in cell

culture media for 24 hours at 37°C, to ensure an 80% confluence. As an un-treated

control, cells were treated with working cell culture medium alone. Cells were also

treated with single agent reovirus at doses representing 0.13, 0.26, 0.44, 0.88, 1.75,

3.51 and 7.02 ×IC50; Cabazitaxel at 0.05, 0.11, 0.22, 0.44, 0.88, 1.75, 2.63 and 3.51

×IC50; or Docetaxel at 0.07, 0.14, 0.29, 0.57, 0.86, 2.86 and 5.71 ×IC50. Additionally,

reovirus was combined with Cabazitaxel or Docetaxel at each dose. Cells were

treated in triplicate for a total of 96 hours at 37°C, before developing the plates with

Cell Titre 96® AQueous One Solution Reagent (Promega, UK), as described in Section

2.9. The % cell survival was calculated relative to untreated cells, before evaluating

the fraction of dead cells affected in each treatment (Section 2.9), which was inputted

into the Bliss Independence analysis spreadsheet [223] (provided by Professor Kevin

Harrington’s laboratory, with permission from MedImmune LLC, USA). Thus, the

Bliss equations [224] provided a powerful statistical model to analyse the in vitro

interaction between reovirus and Cabazitaxel or Docetaxel at doses much lower than

the IC50 of each agent, at non-fixed dose ratios (Section 2.27.6).

67

2.24. ONE STEP GROWTH CURVE ANALYSIS BY THE VIRUS PLAQUE

ASSAY

The plaque assay is a standard method used to determine the infectious virus titre in a

sample, by manually counting the number of plaque forming units (pfu) formed after

infecting a monolayer of host cells. A plaque is formed when a virus infects a host

cell, which will lyse, enabling the virus to spread to and lyse neighbouring cells. It is

assumed that each plaque formed is representative of one infectious viral particle.

The DU145 PCa cell line was treated with reovirus alone at the IC50 dose, or with

reovirus (at the IC50) in combination with Cabazitaxel or Docetaxel at concentrations

representing 1.00, 0.25 and 0.06 ×IC50. Intracellular and extracellular viral samples

were then prepared as described in Section 2.20.1.

2.24.1. Infection of the host-cell monolayer and counting plaques

L929 cells were seeded (1mL per well) in 6-well tissue culture plates at a

concentration of 1x106cells/mL in cell culture medium. Plates were incubated for 24

hours at 37°C to enable cells to reach 90% confluence. 1:10 dilutions of the relevant

reovirus sample was made in working media, ranging from 1×10-1 to a 1×10-9

dilution. The wells were aspirated and replaced with 1mL of each virus dilution in

duplicate, or with 1mL working cell culture media for use as an un-infected control.

The virus was absorbed for 3 hours at 37°C, with intermittent rocking of the plate.

The wells were then aspirated and replaced with 2mL/well warm overlay agar (made

by mixing warm 2% SeaPlaque agarose (Lonza, UK) with an equal volume of

2×MEM (Sigma, UK)). The agar was allowed to set at room temperature for 30

minutes before transferring the plates to the incubator at 37°C for 3 days. Each well

was stained with 1mL Neutral Red solution (Sigma, UK) diluted 1:10 in PBS (×1)

(Fisher Scientific, UK) and incubated at 37°C for 3 hours. The number of plaques

was counted in each well with the aid of a lightbox. The pfu/mL was calculated using

the formula:

pfu/mL = (number of plaques/well) × (volume of sample added in mL) × (stock

viral titre)

68

2.25. INHIBITION OF APOPTOSIS BY z-VAD-FMK

z-VAD-FMK (Sigma, UK) is a competitive, irreversible inhibitor of caspase-1 and

caspase-3-related proteases, and was used to determine whether apoptosis was

contributing to cell death after combination treatment with reovirus and chemotherapy

drugs. The DU145 PCa cell line was seeded (100µL per well) in 96-well plates in cell

culture media for 24 hours at 37°C, to ensure an 80% confluence. After aspirating the

wells, cells were treated with 100µL/well reovirus, Cabazitaxel or Docetaxel as single

agents at doses representing 0.25, 0.5, 1.0, 2.0 and 4.0 ×IC50 values. In addition, cells

were treated with reovirus in combination with Cabazitaxel or Docetaxel at 1.0, 0.50,

0.25, 0.13, 0.06 and 0.03 ×IC50. After 90 minutes, the wells were aspirated and

replaced with 100µL/well 50µM z-VAD-FMK (Sigma, UK) or with 100µL/well

working cell culture media alone as an un-treated control, for 96 hours at 37°C. All

treatments were added in triplicate wells. The plates were developed with the Cell

Titre 96® AQueous One Solution Reagent (Promega, UK) and the % cell survival

relative to un-treated cells was calculated, as described in Section 2.9.

2.26. INHIBITION OF NECROPTOSIS BY NECROSTATIN-1 (NCS-1)

NCS-1 is an inhibitor of necroptosis, and was used to study the mode of cell death

after combination treatment of reovirus and chemotherapy drugs. The DU145 PCa

cell line was seeded (100µL per well) in 96-well plates in cell culture media for 24

hours at 37°C, to ensure an 80% confluence. The wells were aspirated and

100µL/well 30µM NCS-1 (Sigma, UK) or working cell culture media (un-treated

control) was added for 45 minutes and incubated at 37°C. Subsequently, the wells

were aspirated and replaced with 100µL/well reovirus, Cabazitaxel or Docetaxel as

single agents at doses representing 0.25, 0.5, 1.0, 2.0 and 4.0 ×IC50 values. Cells were

also treated with reovirus in combination with Cabazitaxel or Docetaxel at 1.0, 0.50,

0.25, 0.13, 0.06 and 0.03 ×IC50. All treatments were added in triplicate and incubated

at 37°C for 90 minutes. After this time, the plate was aspirated and replaced with

100µL/well cell culture media, and incubated at 37°C for the remaining time totalling

96-hours. The plates were developed with the Cell Titre 96® AQueous One Solution

Reagent (Promega, UK), and the % cell survival relative to un-treated cells was

calculated, as described in Section 2.9.

69

2.27. STATISTICAL ANALYSIS

2.27.1. Significance levels

All statistical calculations were performed using GraphPad Prism Version 6.0

software, unless otherwise specified. For all tests, a 5% significance level was used.

Therefore, the following symbols were used in figures to show the level of

significance:

*p<0.05; **p<0.01; ***p<0.001 and ****p<0.0001

2.27.2. Comparing % cell survival, protein expression or viral titre between

sample means

An un-paired student’s t-test was used to compare statistical differences in % cell

survival between the means of two treatment groups, after subsequent reovirus

infection at a specific MOI. These experiments included siRNA-mediated gene

knock-down in the PJ41 cell line; transient and stable YAP1 over-expression in PJ34

and HN5 cell lines; S1P treatment of the PJ41 cell line; and z-VAD-FMK or NCS-1

treatment of the DU145 cell line. An un-paired student’s t-test was also used to

compare between two means to analyse protein expression levels of JAM-A in

SCCHN cell lines, IFN-β secretion in cell supernatants, and intra- and extracellular

reovirus titre. A one-way ANOVA and Tukey’s post-hoc test was used to compare

statistical differences in vial titre between 3 or more cell lines or treatments. Various

research groups have used these tests for similar purposes [225-231].

2.27.3. Comparing YAP1 protein expression in human tissue samples

The Chi squared (χ) statistical test was used to compare YAP1 positive and negative

IHC staining in head and neck carcinoma or normal tissue cores [232, 233]. It was

also used to compare YAP1 cellular localisation with staining intensity, and to test the

association of YAP1 with clinical factors including tumour grade, tumour stage, age

and sex of the patient.

70

2.27.4. Determining an IC50 value by the median-effect equation

To determine an IC50 value after treatment of a cell line with reovirus or drug, the

median-effect equation of Chou [217] was used via CalcuSyn software (Biosoft, UK).

The median-effect equation is:

Fa/fu = (D/Dm)m

Where D is the dose of drug, Dm is the median-effect dose (IC50), fa is the fraction

affected by the dose, fu is the fraction unaffected (fu = 1-fa) and m is an exponent

indicating the sigmoidicity of the dose effect curve.

2.27.5. The Chou and Talalay equation for measuring the interaction between

two agents at fixed-dose ratios

The interaction of reovirus and Cabazitaxel or Docetaxel in combination was assessed

using CalcuSyn software (BioSoft, UK), which uses the combination index (CI)

equation derived by Chou and Talalay [234]. Generally, a CI of 1 denoted an additive

interaction, >1 antagonism and <1 synergy. The CI equation is:

The denominators of this equation dictate that (Dx)1 and (Dx)2 are the doses of drug 1

and drug 2 alone, respectively, that cause x% cell death. The numerators dictate that

(D)1 and (D)2 are the concentrations of drug 1 and drug 2 in combination that cause

x% cell death. When CI=1, isobolograms can be generated at different effect levels.

For example, ED50, ED75 and ED90 represent the effective dose of two drugs in

combination required to cause 50, 75 and 90% cell death respectively. Combination

data points below the line of additivity on the isobologram plot represent synergism,

whereas data points above the line indicate an antagonistic interaction. Table 2.15

shows the symbols used for describing the interaction in drug combination studies.

(D)1 (D)2 (D)1 (D)2

CI = + +

(Dx)1 (Dx)2 (Dx)1 (Dx)2

71

Table 2.15. Recommended symbols for describing the interaction between reovirus and Cabazitaxel or

Docetaxel when analysed by the CI equation of Chou and Talalay.

Range of CI Symbol Description

<0.1 +++++ Very strong synergism

0.1-0.3 ++++ Strong synergism

0.3-0.7 +++ Synergism

0.7-0.85 ++ Moderate synergism

0.85-0.90 + Slight synergism

0.90-1.10 ± Nearly additive

1.10-1.20 - Slight antagonism

1.20-1.45 -- Moderate antagonism

1.45-3.3 --- Antagonism

3.3-10 ---- Strong antagonism

>10 ----- Very strong antagonism

2.27.6. The Bliss Independence equations for measuring the interaction between

two agents at non-fixed dose ratios

Supposing that two drugs, a and b, both inhibit cancer cell growth, and fractions

affected are Fa and Fb respectively. If two drugs work independently, the combined

inhibition effect can be predicted using the complete additivity probability theory (the

Bliss Independence equation) [223, 224]:

Eexp = (Fa + Fb) – (Fa × Fb)

To illustrate synergy or antagonism from Bliss analysis, a second equation was

applied. If the observed effect (Eobs) was equal to the expected effect (Eexp) (i.e. the

difference in effect (ΔE) and its 95% confidence interval was equal to zero), then the

conclusion was Bliss independence (or addition). If ΔE and its 95% confidence

interval was greater than zero, the combination treatment was thought to be more

efficacious than expected and was synergistic. If ΔE and its 95% confidence interval

was less than zero, the combination treatment was worse than expected and was

antagonistic. The second Bliss equation is [223, 224]:

ΔE = Eobs - Eexp

In the results, for each dose of reovirus and Cabazitaxel or Docetaxel used in

combination, both the ΔE value and upper and lower confidence intervals (Ci) were

presented in a table format, and expressed as a percentage. For example, a 20%

synergistic effect equated to ΔE=0.20. A colour-coded contour map of the ΔE values

was also plotted for easy visualisation of the level of interaction at each combination.

72

CHAPTER 3

TESTING TARGET GENES THAT MAY

INFLUENCE THE SUSCEPTIBILITY OF SCCHN

CELL LINES TO REOVIRUS ONCOLYSIS

73

3. TESTING TARGET GENES THAT MAY INFLUENCE THE

SUSCEPTIBILITY OF SCCHN CELL LINES TO REOVIRUS ONCOLYSIS

3.1. INTRODUCTION

Squamous cell carcinoma of the head and neck (SCCHN) is the most common type of

head and neck cancer. Up to 50% of SCCHN patients present with advanced disease

[20], and standard treatments are surgery combined with chemotherapy and

radiotherapy. However, patients often relapse and develop loco-regional recurrences,

distant metastasis and second primary tumours [5, 21]. These patients have an overall

5-year survival rate of less than 10% [21], a statistic that has not significantly changed

in decades. New treatment strategies are needed to tackle the genetic and biological

heterogeneity of SCCHN, and the resistance that these tumours often develop to

chemotherapy and radiotherapy [5].

Preclinical data has demonstrated reovirus T3D (Reolysin®) to have anti-cancer

activity in SCCHN cell lines [235]. It was also recently shown that HPV-negative

SCCHN cell lines were significantly more susceptible to reovirus oncolysis compared

to HPV-positive SCCHN cell lines. This suggested that reovirus is an appropriate

therapy for HPV-negative SCCHN tumours, which are associated with poor patient

prognosis [236]. A randomized, double-blind Phase III clinical trial is evaluating

overall survival and progression free survival following intravenous (IV)

administration of Reolysin® in combination with paclitaxel and carboplatin versus

chemotherapy alone, in patients with metastatic or recurrent SCCHN [237]. Results

of the primary and secondary endpoints are yet to be published.

Early in vitro studies suggested that reovirus selectively replicates in cells with a

constitutively activated Ras signalling pathway, either through Ras mutation or

upregulated receptor tyrosine kinases, such as the EGFR. This in part provided a

mechanism for the selective oncolysis in tumour cells, as activating mutations in Ras

genes alone have been found in >30% of all human cancers [130, 238]. It would

therefore seem logical to use reovirus in tumours with a high occurrence of Ras

mutations. However, other reports have shown that active Ras signalling alone does

not control the susceptibility to reovirus oncolysis [139], as demonstrated in various

different cancer cell lines [140-144]. Importantly, Twigger et al performed a

74

correlative analysis between reovirus IC50 and EGFR level on a panel of human

SCCHN cell lines with diverse sensitivities to reovirus oncolysis, but found no

association between these two parameters. They also inhibited or stimulated EGFR

signalling and downstream components of Ras, but this was found to have no effect

on reovirus oncolysis [145], nor did pharmacological inhibition of PKR [145]. Ras

signalling may play a role in reovirus oncolysis in certain cancer types, but the full

mechanism of reovirus-induced cancer cell death is more complex and still remains to

be found. Understanding this process may lead to the discovery of predictive

biomarkers of reovirus treatment response that may improve clinical trial design.

Table 3.1 shows the reovirus IC50 dilution values for 9 previously characterised

human SCCHN cell lines [239], as performed by Professor Kevin Harrington’s

laboratory at The Institute of Cancer Research, London [145]. The SCCHN cell lines

showed a broad range of sensitivities to reovirus-induced cell death.

Table 3.1. Professor Kevin Harrington’s laboratory showed that 9 SCCHN cell lines had

different sensitivities to reovirus-induced cell death [145]. SCCHN cell lines were infected with

reovirus at 1.4x109 TCID50/mL diluted 2-fold, starting from a 1:500, a 1:1000 or a 1:5000 dilution.

Cell survival was assessed by an MTS assay at 96 hours post infection. Data were logged transformed

and plotted as sigmoidal dose response curves, with un-infected controls assigned an arbitrary value of

1x108. The IC50 values were interpolated from the dose response curves [145]. The cell lines had a 5-

log range in reovirus IC50 dilution values and were ranked from left-most sensitive to right-most

resistant.

Cell line PJ34 011A O13 Cal 27 SIHN

5B HN3

SIHN

11B HN5 PJ41

Reovirus IC50 dilution

values 1.70e-7

1.80e-7

3.80e-7

1.20e-6

1.50e-5

3.00e-4

>2.00e-3

>2.00e-3

>2.00e-2

Many cancer cell lines are highly sensitive to reovirus oncolysis compared to normal

cells, including NSCLC cells, which displayed reovirus IC50 values of 1.46 to 2.68

log10 pfu/cell at 48 hours post-infection [141]. Pancreatic cancer cell lines infected

with MOI 0.1 virus particles/cell were all relatively sensitive and less than 50% viable

after 2-7 days infection with reovirus [136]. The breast cancer cell lines T47D and

MCF-7 were particularly sensitive to reovirus infection (<20% viability) at MOI 20

75

for 48 hours, as were U87 glioblastoma cells and HepG2 liver cancer cells [144].

However, some cancer cell lines are surprisingly resistant to reovirus-mediated cell

death, including some of the SCCHN cells tested by Twigger et al [145]. For

example, a high percentage of the PJ41 SCCHN cell line population were still viable

after 96 hours infection with reovirus at a concentration of 1.4×107 TCID50/mL [145].

It is difficult to directly compare data in these studies as the experiments were all

performed slightly differently, and thus, we can only estimate the susceptibility to

reovirus in different cancer cell lines.

One approach to studying the host cell response to virus infection is by a technique

called stable isotope labelling by amino acids in cell culture (SILAC). Up- and down-

regulated proteins were examined in HEK293 cells infected with or without reovirus

[240]. This revealed novel host proteome factors that may be involved in virus

infection, including branched chain amino-acid transaminase 2 (BCAT), target of

myb1 (TOM1), paternally expressed 10 (PEG10), tubulin beta 4B (TUBB4B) and

histone cluster 2 H4b (HIST2H4B) [240]. These proteins are good targets for further

analysis and have roles in DNA replication, recombination and repair, as well as

functions in cellular immunology, cell death and survival [240]. Alternatively, gene

expressing profiling using DNA microarrays has been used considerably in cancer

research [241], as it allows the expression of thousands of genes to be measured

simultaneously, in a quick and efficient manner. Two studies have utilised this

method to identify host cell factors that may influence the efficacy of oncolytic

viruses in cancer cell lines [150, 151]. Our laboratory used gene expression profiling

and microarray analysis on a panel of SCCHN cell lines derived from advanced

cancer patient tumours, to detect host-cell factors that may influence their

susceptibility to reovirus oncolysis. Table 3.2 summarises the 8 genes identified in

the microarray screen and the known functions of their corresponding proteins.

Although the function of ZNF600 is generally un-known, the protein products of other

genes, notably SLCO1B3, MGMT, SLC36A4, YAP1 and BIRC2 have been linked to

cancer. The cellular inhibitor of apoptosis-1 (cIAP1) protein (encoded by BIRC2) can

also be proteolytically cleaved and degraded to promote reovirus-induced apoptosis

[242, 243], as discussed in more detail in Chapter 5. Surprisingly, none of the genes

appear to be involved in any known intracellular host-cell anti-viral response

pathway.

76

Table 3.2. Summary of the 8 genes identified as potential predictors of susceptibility to reovirus oncolysis in SCCHN cell lines and their corresponding protein

functions. The genes were identified by Professor Richard Morgan, Oncology Department, University of Surrey (R.Morgan, 2007, unpublished).

Gene name Definition Accession

number Known functions of the corresponding protein References

SLCO1B3

Homo sapiens solute carrier organic

anion transporter family member 1B3.

Synonyms: OATP1B3, LST-3,

HBLRR, LST-2, LST-3TM13, OATP-

8, OATP8 and SLC21A8.

NM_019844

LST-3 is a member of the solute carrier organic anion transporter superfamily, and is

expressed predominantly in the sinusoidal membrane of the liver. It is an important

membrane transport protein that mediates hepatic uptake of endogenous compounds and

environmental toxins. It also mediates uptake of various drugs including anti-cancer agents,

immunosuppressants, lipid-lowering statins and antibiotics. Its Expression has been shown

to be higher in Indocyanine green (IGC)-accumulated hepatocellular carcinoma (HCC) than

in ICG-low HCC. Mutations predicted to cause complete and simultaneous deficiencies of

OATP1B3 has been linked to Rotor syndrome, an autosomal recessive disorder

characterized by conjugated hyperbilirubinemia and coproporphyrinuria.

[244-246]

MGMT Homo sapiens O-6-methylguanine-

DNA methyltransferase NM_002412

MGMT is a DNA repair protein that is involved in cellular defence against mutagenesis and

toxicity. It catalyses the transfer of methyl groups from O(6)-alkylguanine and other

methylated regions of DNA to its own molecule, thus repairing the toxic lesions.

Alkylating agents are used to treat various cancers, but their function has been found to be

limited when MGMT is present. However, if the MGMT promoter region is methylated

then the cells no longer produce MGMT, and are therefore more receptive to alkylating

agents. Methylation of the MGMT gene promoter has been associated with several cancer

types, including glioblastoma, lung cancer and colorectal cancer.

[247]

SLC36A4

Homo sapiens solute carrier family 36

(proton/amino acid symporter),

member 4, transcript variant 1.

Synonyms: PAT4.

NM_152313

The solute carrier (SLC) family are amino acid transporters and are grouped into two main

clusters, the α- and β-families. SLC36 has four members, SLC36 A1-A4. The proteins

from this family are named proton/amino acid transporters (PATs) and are numbered PAT1

to PAT4. PATs have been shown to modulate the activity of mammalian target of

rapamycin complex 1 (mTORC1), which promotes normal growth and is frequently

hyperactivated in tumour cells.

[248, 249]

77

YAP1

Homo sapiens Yes associated protein-

1, transcript variant 2.

Synonyms: COB1, YAP, YAP2,

YAP65 and YKI

NM_006106

YAP1 is a major downstream target protein of the mammalian Hippo signalling pathway, an

evolutionary conserved regulator of cell proliferation and apoptosis. Activation of upstream

serine/threonine kinases phosphorylate down-stream YAP1 on a specific serine residue

(S127) leading to its cytoplasmic retention and inactivation. In the absence of Hippo

signalling, YAP1 migrates to the nucleus of the cell where it interacts with transcription

factors which stimulates expression of genes that promote proliferation. Impairment of

Hippo signalling and nuclear location of YAP1 has been detected in liver, colon, ovarian,

lung and prostate cancers.

[250-256]

ZNF600 Homo sapiens zinc finger protein 600.

Synonyms: KR-ZNF1. NM_198457

The precise function of the ZNF600 protein is unknown, although it has suspected roles in

DNA binding, metal ion binding and transcriptional factor activity. The ZNF600 gene was

recently associated with circulating phospholipid levels.

[257]

P2RY6

Homo sapiens pyrimidinergic receptor

P2Y, G-protein coupled 6, transcript

variant 2

NM_176798

The family of P2 receptors is subdivided into P2X ligand-gated ion channels and P2Y G-

protein coupled receptors, which are activated by extracellular nucleotides. Eight P2Y

receptors have been cloned in humans and are expressed in most epithelia. P2RY6 is a Gq/11

coupled receptor and is responsive to UDP, partially responsive to UTP and ADP, but not

responsive to ATP. Nucleotides released from airway epithelial cells during asthmatic

inflammation activate P2RY6 receptors, leading to release of inflammatory cytokines IL-6

and IL-8, thus regulating immune functions.

[258, 259]

BIRC2

Homo sapiens baculoviral IAP repeat

containing 2, transcript variant 1.

Synonyms: API1, c-IAP1, cIAP1,

Hiap-2, HIAP2, MIHB and RNF48.

NM_001166

The BIRC2 gene encodes a protein that belongs to the inhibitors of apoptosis protein (IAP)

family. These proteins serve as endogenous inhibitors of apoptosis by inhibiting proteases

caspase -3 and -7 and by binding to tumour necrosis factor receptor-associated factors

(TRAF1/2). cIAP1 and cIAP2 have been shown to facilitate cancer cell survival by

functioning as E3 ligases that promote the ubiquitination of receptor interacting protein 1

(RIP1).

[260-262]

LARP1B

Homo sapiens La ribonucleoprotein

domain family member 1B, transcript

variant 2.

Synonyms: LARP2.

NM_178043

The LARP1B gene encodes a protein containing a La motif. La proteins are ubiquitous in

eukaryotic cells and associate with the 3-termini of newly synthesised small RNAs to protect

them from exonucleases, which is required for pre-tRNA maturation. Hypermethylation of

the LARP1B gene may be epigenetically inherited, and has the potential to be used as a

marker for early prenatal diagnosis of -thalassemia.

[263, 264]

78

3.2. STUDY OBJECTIVE

The objective of this chapter was to determine whether any of the 8 target genes

identified in the DNA microarray screen, influenced the susceptibility of human

SCCHN cell lines to reovirus-mediated cell death.

In order to test this, the following experiments were performed:

1. Infection of 3 representative SCCHN cell lines with reovirus and assessment

of oncolysis by the Cell Titer 96® AQueous Non-radioactive Cell Proliferation

Assay (MTS assay), to validate previous findings.

2. Infection of non-cancerous, un-transformed human cells with reovirus and

assessment of oncolysis by the MTS assay, to validate the use of reovirus as an

anti-cancer agent.

3. Measurement of the mRNA expression levels of the 8 target genes in 3

representative SCCHN cell lines by RT-qPCR, to validation previous findings.

4. siRNA-mediated knock-down of the 8 genes in the reovirus-resistant PJ41

SCCHN cell line by lipid transfection, and measurement of the mRNA

expression by RT-qPCR to ensure efficient knock-down of the genes.

5. siRNA-mediated knock-down of the 8 target genes in the PJ41 SCCHN cell

line and subsequent infection of the cells with reovirus, to assess the efficiency

of reovirus oncolysis by the MTS assay.

6. Assessment of the efficiency of siRNA-mediated knock-down at the protein

level in the PJ41 SCCHN cell line by western blotting.

79

3.3. RESULTS

3.3.1. Host cell mRNA expression of 8 genes directly correlated with reovirus

IC50 in SCCHN cell lines

In order to identify host cell factors that may influence the susceptibility to reovirus

oncolysis, gene expression profiling was carried out on 5 of the SCCHN cell lines

using microarray hybridisation, via the Low Impact Quick Amp Labelling Kit, one-

color (Agilent Technologies, USA). PJ34, O11A and PJ41 cell lines were analysed in

duplicate, whereas O13 and SIHN 11B were analysed as single samples. A total of

44,000 gene probes were tested and the extracted data was subsequently analysed

using Genespring software (Agilent Technologies, USA). Results showed that the

expression of 8 gene probes increased as the cell lines became progressively more

resistant to reovirus-induced cell death. The mRNA expression of the 8 genes

(SLCO1B3, MGMT, SLC36A4, YAP1, ZNF600, P2RY6, BIRC2 and LARP1B)

identified in the microarray was then measured by RT-qPCR in all 9 SCCHN cell

lines (PJ34, O11A, O13, Cal27, SIHN 5B, HN3, SIHN 11B, HN5 and PJ41). The

same correlation was observed, i.e. as the resistance to reovirus oncolysis increased

across the panel of cell lines, the mRNA expression of all 8 genes also generally

increased (Figure 3.1). This research was performed in-house by Professor Richard

Morgan, Oncology Department, University of Surrey (R.Morgan, 2007, unpublished).

The accession number of these genes and the known functions of their corresponding

proteins are summarised in Table 3.2. The correlation coefficient (r value) between

the relative mRNA expression of the 8 genes and the reovirus IC50 value (Table 3.1)

in the cell lines was computed, which confirmed a positive trend (Figure 3.2).

80

Figure 3.1. RT-qPCR analysis of microarray-identified up-regulated genes in 9 SCCHN cell lines. cDNA from PJ34, 011A, 013, Cal27, SIHN 5B, HN3,

SIHN 11B, HN5 and PJ41 cell lines was analysed by RT-qPCR. The mRNA expression of 8 genes (SLCO1B3 (dark pink bars), MGMT (blue bars), SLC36A4 (green

bars), YAP1 (yellow bars), ZNF600 (orange bars), P2RY6 (light pink bars), BIRC2 (grey bars) and LARP1B (black bars)) is shown on a log10 scale relative to the

housekeeping gene β-actin (×1000). As the mRNA expression of these genes increased across the panel of SCCHN cell lines, so did the resistance to reovirus

oncolysis (R.Morgan, 2007, unpublished). The graph represents the mean of duplicate samples.

PJ

34

01

1A

O1

3

Ca

l27

SIH

N 5

BH

N3

SIH

N 1

1B

HN

5

PJ

41

0 .0 1

0 .1

1

1 0

1 0 0

Re

lati

ve

mR

NA

ex

pre

ss

ion

S L C O 1 B 3

M G M T

S L C 3 6 A 4

Y A P 1

Z N F 6 0 0

P 2 R Y 6

B IR C 2

L A R P 1 B

81

Figure 3.2. Pearson correlation coefficient between the relative mRNA expression of the 8 genes and the reovirus IC50 dilution value in SCCHN cell lines.

The mRNA expression of A. SLCO1B3, B. MGMT, C. SLC36A4, D. YAP1, E. ZNF600, F. P2RY6, G. BIRC2 and H. LARP1B, was plotted against the reovirus IC50

dilution value in each cell line. PJ34, 011A, 013, Cal27, SIHN 5B, HN3, SIHN 11B, HN5 and PJ41 cell lines are represented as black circles. The correlation

coefficient (r value) is shown in red on each graph, and all showed a positive trend. The data are shown on a log10 scale.

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .0 1

0 .1

1

1 0

S L C O 1 B 3

R e o v iru s IC 5 0 d ilu tio n v a lu e

Re

lati

ve

mR

NA

ex

pre

ss

ion

P J 3 4

0 1 1 A

O 1 3

C a l2 7

H N 3 S IH N 1 1 B

H N 5

P J 4 1r = 0 .8 3 8 1

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .1

1

1 0

M G M T

R e o v iru s IC 5 0 d ilu tio n v a lu eR

ela

tiv

e m

RN

A e

xp

re

ss

ion

O 1 3

C a l2 7

S IH N 5 B

H N 3

S IH N 1 1 B

H N 5

P J 4 1

r = 0 .1 9 6 2

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .0 1

0 .1

1

1 0

Z N F 6 0 0


Re

lati

ve

mR

NA

ex

pre

ss

ion

P J 3 4

0 1 1 A

O 1 3

C a l2 7

S IH N 5 B H N 3

S IH N 1 1 B

H N 5

P J 4 1r = 0 .8 1 0 0

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .1

1

1 0

1 0 0

P 2 R Y 6


Re

lati

ve

mR

NA

ex

pre

ss

ion

P J 3 4

0 1 1 A

O 1 3

C a l2 7

S IH N 5 B

H N 3

S IH N 1 1 B

H N 5

P J 4 1

r = 0 .9 8 5 9

A B

E F

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .0 1

0 .1

1

1 0

S L C 3 6 A 4


Re

lati

ve

mR

NA

ex

pre

ss

ion

P J 3 4

0 1 1 A

O 1 3C a l2 7

S IH N 5 B

H N 3S IH N 1 1 BH N 5

P J 4 1

r = 0 .8 5 6 1

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .1

1

1 0

1 0 0

Y A P 1


Re

lati

ve

mR

NA

ex

pre

ss

ion

P J 3 4

0 1 1 AO 1 3C a l2 7

S IH N 5 B

H N 3

S IH N 1 1 BH N 5

P J 4 1

r = 0 .9 9 6 1

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .1

1

1 0

B IR C 2


Re

lati

ve

mR

NA

ex

pre

ss

ion

P J 3 4

0 1 1 AO 1 3

C a l2 7

S IH N 5 B

H N 3 S IH N 1 1 B

H N 5

P J 4 1r = 0 .9 8 3 7

1 0 -7 1 0 -6 1 0 -5 1 0 -4 1 0 -3 1 0 -2 1 0 -1

0 .1

1

1 0

L A R P 1 B


Re

lati

ve

mR

NA

ex

pre

ss

ion

P J 3 4

0 1 1 A

O 1 3

C a l2 7S IH N 5 B

H N 3

S IH N 1 1 B

H N 5

P J 4 1

r = 0 .7 8 6 1

C D

G H

82

3.3.2. Validation of SCCHN cell line susceptibility to reovirus-induced cell death

The first step in this project was to validate the reovirus IC50 data published by

Twigger et al (Professor Kevin Harrington’s laboratory) [145]. First, the reovirus

stock was titred by a standard plaque assay on the highly sensitive L929 mouse

fibroblast cell line. Using this stock titre, the amount of virus needed to kill 50% of

different cell populations (IC50 dilution) could be calculated and compared. The

experiment from the Twigger et al paper [145] was then performed on 3 of the

SCCHN cell lines; PJ34, HN5 and PJ41. These cell lines were chosen because their

IC50 values represented low, medium and high levels of resistance to reovirus-induced

cell death (1.7e-7, >2.00e-3 and >2.00e-2 TCID50/mL respectively, as shown in Table

3.1). The cells lines were infected with serial dilutions of reovirus T3D for 24 or 48

hours and then the % cell viability was calculated using the MTS assay (Section 2.9).

The IC50 values were calculated using CalcuSyn software (Biosoft, UK) (Section

2.27.4). Results were consistent with the findings produced by Twigger et al. PJ34,

HN5 and PJ41 still represented SCCHN cell lines that had low, medium and high

resistance to reovirus oncolysis (IC50s were MOI 572.6, 29.5 and 6.7 at 24 hours post

infection (hpi) and MOI 248.5, 11.3 and 2.7 at 48hpi respectively) (Figure 3.3 and

Table 3.3).

It has been extensively documented that reovirus T3D has a natural propensity to

infect and lyse various different cancerous or transformed cells. In order to validate

the reovirus stock in our laboratory, 2 representative normal cell types were used. A

normal human head and neck cell line was not available. Therefore an untransformed

normal human lung fibroblast cell line, MRC-5, and PBMCs isolated from the blood

of a healthy human donor (Section 2.21.1), were infected with reovirus T3D at

various MOIs for 24 or 48 hours. The % cell survival was then assessed by the MTS

assay (Section 2.9) and the IC50 values were calculated (Section 2.27.4). As

expected, both MRC-5 and PBMCs showed high levels of resistance to reovirus

induced cell death compared to SCCHN cell lines (IC50s were MOI 8040.6, 2094.9 at

24hpi and MOI 2769.7, 1509.6 at 48hpi respectively) (Figure 3.3 and Table 3.3).

This confirmed that the reovirus T3D stock was cancer-cell specific and suitable for

this study.

83

A

B

Figure 3.3. Validation of the susceptibility to reovirus oncolysis in 3 SCCHN cell lines and in 2

non-cancerous, untransformed cell types. PJ34 (yellow triangles), HN5 (orange squares), PJ41 (red

circles), PBMC (open circles) and MRC-5 (open triangles) cells, were infected with serial dilutions of

reovirus, starting at a multiplicity of infection (MOI) 1000 or 4000 for A. 24 hours or B. 48 hours. The

% cell survival was measured using the MTS assay. SCCHN cell lines PJ34, HN5 and PJ41

represented low, medium and high resistance to reovirus-induced cell death respectively. MRC-5 and

PBMC (both non-cancerous, untransformed cell types) were considerably more resistant to reovirus

oncolysis than the SCCHN cell lines. The graphs show the mean of 3 assay repeats and error bars

represent SEM.

0.0

2.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

2000.0

4000.0

0

2 5

5 0

7 5

1 0 0

1 2 5

4 8 h o u r s

R e o v iru s (M O I)

% c

ell

su

rv

iva

l

P B M C

M R C -5

P J 4 1

H N 5

P J 3 4

0.0

2.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

2000.0

4000.0

0

2 5

5 0

7 5

1 0 0

1 2 5

2 4 h o u r s


% c

ell

su

rv

iva

l P B M C

M R C -5

P J 4 1

H N 5

P J 3 4

84

Table 3.3. Reovirus IC50 values of 3 SCCHN cell lines, the MRC-5 human fibroblast cell line and

PBMCs isolated from a healthy donor, were calculated using CalcuSyn software. The values

represent the mean IC50 of 3 independent experiments ± SEM. The cell lines were ranked from left-

most sensitive to right-most resistant to reovirus oncolysis.

Cell type PJ34 HN5 PJ41 PBMC MRC-5

IC50 of reovirus

(MOI) at 24

hours ± SEM

6.7 ± 0.781 29.5 ± 1.600 572.6 ± 76.074 2094.9 ± 169.034 8040.6 ± 221.003

IC50 of reovirus

(MOI) at 48

hours ± SEM

2.7 ± 0.470 11.3 ± 1.155 248.5 ± 13.119 1509.6 ± 90.204 2769.7 ± 316.300

Oncolysis levels

85

3.3.3. Validation of mRNA expression of the 8 genes in SCCHN cell lines

In order to study the potential role of the 8 genes identified by Professor Richard

Morgan in reovirus oncolysis, we intended to repeat the RT-qPCR experiment to

clarify whether expression of these genes in SCCHN cell lines is reproducible. The

same 3 cell lines were chosen for analysis; PJ34, HN5 and PJ41. This was because

they previously displayed low, medium and high expression of the 8 target genes

respectively (Figure 3.1). The cDNA template from each cell line was used to

quantify SLCO1B3, MGMT, SLC36A4, YAP1, ZNF600, P2RY6, BIRC2 and LARP1B

expression via RT-qPCR. The relative expression value was calculated as a ratio to

the housekeeping gene β-actin for 2 independent experiments (Sections 2.10, 2.11

and 2.12).

Evaluation of the 3 SCCHN cell lines revealed the same pattern of expression of these

genes, i.e. PJ34<HN5<PJ41 (Figure 3.4). The only exception to this was LARP1B, as

its expression in the HN5 cell line was marginally higher than in PJ41s. The melting

curves for each target gene displayed a single peak, suggesting that the desired

amplicon was detected in the PCR reaction. Having validated Professor Morgan’s

work in 3 SCCHN cell lines, we were confident that our hypothesis was justified; that

7 out of the 8 genes could potentially influence the susceptibility to reovirus

oncolysis.

86

Figure 3.4. mRNA expression validation of the 8 genes in 3 SCCHN cell lines. cDNA from PJ34,

HN5 and PJ41 cell lines was analysed by RT-qPCR. The mRNA expression of SLCO1B3 (dark pink

bars), MGMT (blue bars), SLC36A4 (green bars), YAP1 (yellow bars), ZNF600 (orange bars), P2RY6

(light pink bars), BIRC2 (grey bars) and LARP1B (black bars) is shown on a log10 scale relative to the

housekeeping gene β-actin (×1000). The pattern of expression of these genes was generally

PJ34<HN5<PJ41, which was consistent with Professor Richard Morgan’s data. As the mRNA

expression of these genes increased in the SCCHN cell lines, so did the resistance to reovirus

oncolysis. Error bars represent the SEM from two assay repeats.

PJ34

HN

5

PJ41

0 .0 0 1

0 .0 1

0 .1

1

1 0

1 0 0

1 0 0 0

1 0 0 0 0

Re

lati

ve

mR

NA

ex

pre

ss

ion

S L C O 1 B 3

M G M T

S L C 3 6 A 4

Y A P 1

Z N F 6 0 0

P 2 R Y 6

B IR C 2

L A R P 1 B

87

3.3.4. Optimisation of siRNA-transfection conditions in the PJ41 reovirus-

resistant cell line using the KDalert™ GAPDH assay kit

To study the possible role of the 8 genes (SLCO1B3, MGMT, SLC36A4, YAP1,

ZNF600, P2RY6, BIRC2 and LARP1B) on reovirus-oncolysis, an siRNA screen was

performed in the PJ41 SCCHN cell line. This cell line was chosen because it was the

most resistant to reovirus-induced cell death and exhibited the highest expression of 7

out of the 8 genes. The LARP1B gene was included in this screen, even though its

expression was lower than expected in this cell line.

First, the transfection conditions were optimised in the PJ41 cell line using the

KDalert™ GAPDH assay kit, which is an indirect method of determining GAPDH

knock-down at the protein level (Section 2.13.1). The efficiency of siRNA delivery

can be monitored by measuring the reduction in GAPDH in cells transfected with

GAPDH siRNA relative to cells transfected with a negative control siRNA. The kit

also serves as a marker for identifying cellular toxicity resulting from transfection.

The efficiency of siRNA transfection is strongly influenced by the concentration of

transfection reagent and the cell seeding density used. Therefore, PJ41 cells were

seeded at 2 different concentrations; 8.0×103 and 1.2×104 cells/well. 3 different

concentrations of siPORT Neo FX transfection agent (0.2, 0.5 and 0.8µL/well) were

used to deliver GAPDH or negative control siRNAs. 48 hours after initial incubation

with the siRNAs, the KDalert™ GAPDH assay was performed.

The greatest % GAPDH knock-down was achieved using 8.0×103 cells/well and

0.8µL/well siPORT Neo FX (Figure 3.5 A). However taking into account both the %

GAPDH knock-down and the transfection-associated toxicity, the optimal conditions

were 1.2×104 cells/well and 0.5µL/well siPORT Neo FX. This was determined by the

Optimal Balance Factor (OBF), which multiplied the GAPDH activity by the %

GAPDH knock-down in the samples transfected with negative control siRNA

(Section 2.13.1). The optimal transfection conditions were those which show the

greatest OBF value (Figure 3.5 B).

88

A

B

Figure 3.5. Optimisation of siRNA-mediated transfection conditions in the PJ41 cell line by the

KDalert™ GAPDH assay. A. Shows the % GAPDH knock-down using 0.2, 0.5 and 0.8µL/well Neo

FX transfection agent and 2 cell seeding densities, 8000 and 12000 cells/well. Error bars represent the

SD from triplicate samples. B. Optimal Balance Factor (OBF) was calculated, which makes a

compromise between the % knock-down efficiency and the toxicity of the transfection. OBF multiplies

the GAPDH activity by the % GAPDH knock-down in the samples transfected with negative control

siRNA. The optimal conditions for transfection were 12000 cells/well and 0.5µL/well Neo FX, as this

displayed the greatest OBF value.

89

3.3.5. siRNA-mediated knock-down of the 8 target genes in the PJ41 cell line

Having optimised the transfection conditions, the PJ41 cell line was transiently

transfected with 2 different siRNAs for each of the 8 target genes, as well as a

negative control siRNA. As positive controls, cells were treated with the siPORT Neo

FX transfection agent alone or with media alone (un-transfected sample) (Section

2.13.2). At 48 hours post-transfection, RNA was extracted from the cells and the

cDNA template was used to quantify SLCO1B3, MGMT, SLC36A4, YAP1, ZNF600,

P2RY6, BIRC2 and LARP1B mRNA expression by RT-qPCR. The relative

expression value was calculated as a ratio to the housekeeping gene β-actin (Section

2.12).

Figure 3.6 displays the reduction in relative mRNA expression of each independent

target gene using the 2 different siRNAs. For each target gene, the siRNA that

achieved the greatest reduction compared to the negative control siRNA-treated cells

was chosen for future experiments. Knock-down of all genes using the most effective

siRNA ranged from 79.1 to 99.9% (Table 3.4).

90

SL

CO

1B

3 s

iRN

A (

s26261)

SL

CO

1B

3 s

iRN

A (

s26262)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

2 0

4 0

6 0

Re

lati

ve

mR

NA

ex

pre

ss

ion

MG

MT

siR

NA

(s8750)

MG

MT

siR

NA

(s8752)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0 .0

0 .5

1 .0

1 .5

Re

lati

ve

mR

NA

ex

pre

ss

ion

SL

C36A

4 s

iRN

A (

s42350)

SL

C36A

4 s

iRN

A (

s42351)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

1 0

2 0

3 0

4 0

5 0

Re

lati

ve

mR

NA

ex

pre

ss

ion

YA

P1 s

iRN

A (

s20366)

YA

P1 s

iRN

A (

s20368)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

Re

lati

ve

mR

NA

ex

pre

ss

ion

A

DC

B

91

Figure 3.6. mRNA expression of the 8 target genes in the PJ41 cell line after siRNA-mediated

knock-down. The mRNA expression of A. SLCO1B3, B. MGMT, C. SLC36A4, D. YAP1, E. ZNF600,

F. P2RY6, G. BIRC2 and H. LARP1B, was assessed in cells treated with 2 target siRNAs, negative

control siRNA, Neo FX transfection agent alone or media only, by RT-qPCR. The % knock-down of

each target gene was very efficient and ranged from 79.1 to 99.9% compared to the negative control

siRNA. The mRNA expression of each target gene is shown relative to the housekeeping gene β-actin

(×1000). The graphs show the mean of 2 assay repeats and error bars represent SEM.

ZN

F600 s

iRN

A (

s46376)

ZN

F600 s

iRN

A (

s46378)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

1 0

2 0

3 0R

ela

tiv

e m

RN

A e

xp

re

ss

ion

P2R

Y6 s

iRN

A (

sc-4

2584)

P2R

Y6 s

iRN

A (

s224151)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0 .0

0 .3

0 .6

0 .9

1 .2

Re

lati

ve

mR

NA

ex

pre

ss

ion

BIR

C2 s

iRN

A (

s1448)

BIR

C2 s

iRN

A (

s1449)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

2 0 0

4 0 0

6 0 0

Re

lati

ve

mR

NA

ex

pre

ss

ion

LA

RP

1B

siR

NA

(s30246)

LA

RP

1B

siR

NA

(s30247)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

3

6

9

1 2

Re

lati

ve

mR

NA

ex

pre

ss

ion

E

HG

F

92

Table 3.4. A summary of the % knock-down of each target gene in the PJ41 cell line by

transfection with 2 different siRNAs. siRNAs highlighted in bold achieved the greatest reduction in

relative mRNA expression.

Target gene siRNA ID % knock-down of target gene

SLCO1B3 s26261

79.1

s26262 77.7

MGMT s8750

51.6

s8752 94.5

SLC36A4 s42350

90.5

s42351 97.7

YAP1 s20366

95.9

s20368 96.4

ZNF600 s46376

99.9

s46378 93.7

P2RY6 sc-42584

62.5

s224151 94.2

BIRC2 s1448

97.8

s1449 91.3

LARP1B s30246

83.6

s30247 85.4

93

3.3.6. siRNA-mediated knock-down of YAP1 sensitised the PJ41 cell line to

reovirus-induced cell death

Having achieved successful knock-down of the 8 target genes, the plan was to

evaluate the sensitivity of these cells to reovirus treatment, in comparison to negative

and positive controls (Section 2.13.3). PJ41 cells were transiently transfected with

the siRNA that caused the greatest decrease in mRNA expression of the 8 target genes

(Table 3.4). Cells were also transfected with a negative control siRNA to assess the

non-specific effect caused by scrambled sequences. Cells treated with siPORT Neo

FX transfection agent alone or with media alone served as the positive controls. After

48 hours post-transfection, to account for any differences in transfection-associated

cytotoxicity, viable cells from each treatment condition were counted. The cell counts

were used to calculate the required MOI per well (Section 2.8), which ensured that

the same number of virus particles per cell was used for each treatment condition.

Cells were then subsequently infected with serial dilutions of reovirus starting at MOI

1000. At 48 hours post-infection with reovirus, the % cell survival in each treatment

condition was assessed by the MTS assay (Section 2.9).

In all experiments, the reovirus IC50 value of the negative siRNA treated cells was

somewhat lower than the IC50 obtained for the transfection agent alone treated cells,

suggesting that there was some non-specific sensitisation to reovirus treatment.

Therefore, in order to assess the specific effect of reduced target gene expression, all

comparisons were made directly to the negative control siRNA-treated cells.

Compared to negative control siRNA-treated cells, knock-down of SLCO1B3,

MGMT, SLC36A4, ZNF600, P2RY6, BIRC2 and LARP1B had minimal effect on

reovirus-induced cell death (Figure 3.7 A, B, C, E, F, G and H respectively).

Although there were statistically significant differences at some reovirus

concentrations tested, there was considerable over-lap with the cell survival curves of

the controls. However, Figure 3.7 D shows that siRNA-mediated knock-down of

YAP1 (which encodes the Yes-associated protein-1 (YAP1) protein) caused

significant sensitisation to reovirus-induced cell death at all reovirus MOIs tested

(p<0.05 by un-paired t-test), apart from MOI 1000. YAP1 knock-down cells were

greater than 3-fold more sensitive to reovirus oncolysis than the negative siRNA

control cells (IC50 MOI 115.9 and MOI 385.7 respectively), and there was a clear

94

separation in the % cell survival curves. This suggested that reduced YAP1

expression in a resistant SCCHN cell line may contribute to sensitisation to reovirus-

induced cell death.

The average optical density (OD) of each un-infected transfection condition was also

compared. After 48 hours post-transfection, cells were treated with media alone for a

further 48 hours before analysis via the MTS assay (Section 2.9). There was no

noticeable differences in OD values between cells treated with negative control

siRNA and the target-gene siRNAs (Figure 3.8). This was also reflected in the cell

counts, which were very similar for each treatment condition. This implied that the

observed differences in reovirus sensitivity was not due to differences in transfection-

associated cytotoxicity, but was indeed caused by the reduced expression of the target

genes. Thus, as YAP1 was the only gene that showed sensitisation to reovirus

treatment upon knock-down, it was considered an important host-cell factor and was

selected for further investigation.

95

0.0

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

S L C O 1 B 3 s iR N A

N e g a tiv e s iR N A

N e o F X o n ly

M e d ia o n ly

*

Condition Reovirus IC 50 (MOI)

SLCO1B3 siRNA 534.4 ± 58.8

Negative siRNA 385.7 ± 34.7

Neo FX only 702.6 ± 56.2

Media only 812.9 ± 73.2

0.0

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

M G M T s iR N A


N e o F X o n ly

M e d ia o n ly

*


MGMT siRNA 74.4 ± 8.2


Neo FX only 235.3 ± 18.8

Media only 521.6 ± 15.6

A

B

96

0.0

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

S L C 3 6 A 4 s iR N A


N e o F X o n ly

M e d ia o n ly

**

*

*

*


SLC36A4 siRNA 190.1 ± 22.8


Neo FX only 235.3 ± 18.8

Media only 521.6 ± 15.6

0.0

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

Y A P 1 s iR N A


N e o F X o n ly

M e d ia o n ly

*** ***

** **

*****


YAP1 siRNA 115.9 ± 7.0


Neo FX only 702.6 ± 56.2

Media only 812.9 ± 73.2

C

D

97

0.0

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

Z N F 6 0 0 s iR N A


N e o F X o n ly

M e d ia o n ly

*


ZNF600 siRNA 88.2 ± 3.5


Neo FX only 235.3 ± 18.8

Media only 521.6 ± 15.6

0.0

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

P 2 R Y 6 s iR N A


N e o F X o n ly

M e d ia o n ly

*

**


P2RY6 siRNA 167.3 ± 10.0


Neo FX only 235.3 ± 18.8

Media only 521.6 ± 15.6

E

F

98

0.0

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

B IR C 2 s iR N A


N e o F X o n ly

M e d ia o n ly

*


BIRC2 siRNA 485.5 ± 27.5


Neo FX only 702.6 ± 56.2

Media only 812.9 ± 73.2

0.0

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

L A R P 1 B s iR N A


N e o F X o n ly

M e d ia o n ly

*

**


LARP1B siRNA 1588.3 ± 285.9


Neo FX only 702.6 ± 56.2

Media only 812.9 ± 73.2

G

H

99

Figure 3.7. Evaluation of reovirus-induced cell death after siRNA-mediated knock-down of the 8

target genes in the PJ41 cell line. Cells were transfected with specific siRNAs (red circles) for A.

SLCO1B3, B. MGMT, C. SLC36A4, D. YAP1, E. ZNF600, F. P2RY6, G. BIRC2 and H. LARP1B, and

then subsequently infected with serial dilutions of reovirus, starting at MOI 1000. Cells were also

transfected with a negative control siRNA (purple squares), Neo FX transfection agent alone (green

triangles) or treated with media only (blue triangles). The % cell survival in each treatment condition

was then assessed using the MTS assay. The IC50 values of each treatment condition were determined

using CalcuSyn software and are shown in a table below each graph. Out of the 8 genes tested, knock-

down of YAP1 had the most discernible effect compared to negative control siRNA treated cells, and

sensitised cells to reovirus-induced cell death at nearly all MOIs tested. *p<0.05, **p<0.01,

***p<0.001 and ****p<0.0001 by un-paired t-test, with respect to the negative control siRNA. Error

bars represent the SD from 2 assay repeats.

Figure 3.8. The transfection-associated toxicity in each treatment condition. After 48 hours post-

transfection, PJ41 cells were treated with media alone for a further 48 hours before analysis via the

MTS assay. The raw Optical Density (OD) values for each treatment were compared. There were no

major differences in OD between cells transfected with negative control siRNA and the cells

transfected with each target siRNA, suggesting that transfection-associated cytotoxicity was not a

concern. The graph shows the mean OD of 2 assay repeats and error bars represent SD.

Med

ia o

nly

Neo

FX

on

ly

Neg

ati

ve s

iRN

A

SL

CO

1B

3 s

iRN

A

MG

MT

siR

NA

SL

C36A

4 s

iRN

A

YA

P1 s

iRN

A

ZN

F600 s

iRN

A

P2R

Y6 s

iRN

A

BIR

C2 s

iRN

A

LA

RP

1B

siR

NA

0 .0

0 .5

1 .0

1 .5

2 .0

T ra n s fe c t io n c o n d it io n

Op

tic

al

De

ns

ity

(4

90

nm

)

100

3.3.7. siRNA-mediated knock-down of the YAP1 protein in the PJ41 cell line

Next, having identified YAP1 as an important target of reovirus resistance at the

mRNA level, siRNA-mediated knock-down of the Yes-Associated-Protein-1 (YAP1)

protein in the PJ41 cell line was also determined. Cells were treated with 2 different

YAP1 siRNAs (ID: s20366 and s20368), negative control siRNA, Neo FX

transfection agent alone, or media alone (Section 2.13.2). At 48 hours post-

transfection, cell lysates were collected and YAP1 protein expression was determined

by western blotting (Section 2.14). Transfection with siRNA s20368 caused the

greatest reduction in the YAP1 54kDa band compared to the positive control lysates

(Figure 3.9 A). To ensure uniform protein loading, the blot was re-probed with a β-

actin loading control (Section 2.14.3). The intensity of the YAP1 band in each

sample was quantified and normalised to their corresponding β-actin bands by

densitometry analysis (Section 2.14.4) (Figure 3.9 B). Despite a 96.4% reduction in

YAP1 mRNA expression, a modest 58.4% reduction of the YAP1 protein was

achieved compared to negative control cells. Efforts were made to improve this

reduction by altering the parameters of the transfection assay, but this did not improve

the knock-down. This emphasises the difficulty in achieving complete attenuation of

the YAP1 protein in the PJ41 cell line. However, the reduction of the YAP1 protein

was substantial enough to observe a 3-fold decrease in the reovirus IC50 value

compared to the negative control (Figure 3.7 D).

101

Figure 3.9. YAP1 protein detection in the PJ41 cell line after YAP1 siRNA-mediated knock-

down. Cells were transfected with 2 different YAP1 siRNAs (s20368 and s20366), or with negative

control siRNA. Cells were also treated with Neo FX transfection agent alone or media alone, which

served as additional positive controls for the YAP1 protein. A. Cell lysates were collected and YAP1

protein expression was detected by western blotting. siRNA s20368 caused the greatest decrease in

YAP1 protein expression compared to the negative control siRNA, Neo FX alone and media only

treated cells. B. YAP1 protein expression in each sample was normalised to the -actin loading

control and quantified by densitometry using LI-COR Image Studio Lite 4.0.21 software. There was a

58.4% reduction in YAP1 compared to negative control siRNA treated cells.

YAP1

siR

NA

s20

366

YAP1

siR

NA

s20

368

Neg

ativ

e si

RN

A

Neo

FX

onl

y

Med

ia o

nly

YAP

-1 s

iRN

AS

20

36

8

YAP

-1 s

iRN

AS

20

36

6

Ne

gsi

RN

A

Me

dia

+ s

iPO

RT

Ne

oFX

Me

dia

on

ly

YAP-1 (54KDa)

β-actin (42KDa)

YA

P1 s

iRN

A s

20366

YA

P1 s

iRN

A s

20368

Neg

ati

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

5 0

1 0 0

1 5 0

2 0 0

2 5 0

Re

lati

ve

de

ns

ity

A B

YAP1

siR

NA

s20

366

YAP1

siR

NA

s20

368

Neg

ativ

e si

RN

A

Neo

FX

onl

y

Med

ia o

nly

YAP

-1 s

iRN

AS

20

36

8

YAP

-1 s

iRN

AS

20

36

6

Ne

gsi

RN

A

Me

dia

+ s

iPO

RT

Ne

oFX

Me

dia

on

ly

YAP-1 (54KDa)

β-actin (42KDa)

YA

P1 s

iRN

A s

20366

YA

P1 s

iRN

A s

20368

Neg

ati

ve s

iRN

A

Neo

FX

on

ly

Med

ia o

nly

0

5 0

1 0 0

1 5 0

2 0 0

2 5 0

Re

lati

ve

de

ns

ity

A B

102

3.4. DISCUSSION

Previously published work by Twigger et al showed that several SCCHN cell lines

had diverse sensitivities to reovirus-mediated cell death [145]. Furthermore, gene

expression profiling and RT-qPCR analysis performed by Professor Richard Morgan,

revealed that expression of 8 genes increased as the SCCHN cell lines became

progressively more resistant to reovirus oncolysis. These genes were SLCO1B3,

MGMT, SLC36A4, YAP1, ZNF600, P2RY6, BIRC2 and LARP1B. The aim of this

chapter was to determine whether any of these genes are essentially host cell factors

that influence the susceptibility to reovirus oncolysis. In order to address this, the

mRNA expression profile of these genes and reovirus IC50 values in 3 of the SCCHN

cell lines, was reproduced. Additionally, individual knock-down of the 8 genes and

the resulting effect on reovirus oncolysis was assessed in the PJ41 cell line, as this cell

line was the most resistant to reovirus and generally displayed the highest expression

of the target genes.

Our findings showed that PJ34, HN5 and PJ41 represented SCCHN cell lines that had

low, medium and high resistance to reovirus oncolysis respectively, and concurred

with the data published by Twigger et al [145]. These cell lines were derived from

the primary tumours of advanced SCCHN patients. It is not surprising that these cell

lines have such variable susceptibilities to reovirus treatment, as SCCHN tumours are

known to be heterogeneous and often present with genetic mutations in p53, pRb,

EGFR, TGF-β, and P13K-PTEN-Akt signalling pathways [5]. Furthermore, there are

two distinct groups of SCCHN patients; HPV-negative and HPV positive. HPV-

negative SCCHN tumours have frequent p53 mutations, normally occur in patients

over the age of 60 and have a poor prognosis. HPV-positive SCCHN tumours on the

other hand have infrequent p53 mutations and generally affect younger people who

have a favourable prognosis [5]. HPV-negative SCCHN cell lines were recently

shown to be significantly more susceptible to reovirus oncolysis compared to HPV-

positive SCCHN cell lines, suggesting that reovirus is an appropriate therapy for

HPV-negative head and neck cancers [236]. The HPV-status of the SCCHN cell lines

used in our study is unknown. It would therefore be interesting to determine whether

HPV-status correlates with their reovirus susceptibilities, and is something to consider

for future work.

103

We found that PJ34, HN5 and PJ41 cell lines displayed in turn, low, medium and high

mRNA expression of SLCO1B3, MGMT, SLC36A4, YAP1, ZNF600, P2RY6, BIRC2,

which is in agreement with the preliminary work implemented by Professor Richard

Morgan (R.Morgan, 2007, unpublished). However, we found that LARP1B

expression was slightly higher in HN5 than PJ41, which did not correspond to

previous findings, but was still lowest in PJ34. As the preliminary experiment shown

in Figure 3.1 was conducted several years ago, the cell lines used to validate this data

at the beginning of this project were from different stocks, and were either later

provided to us by another laboratory or bought-in from an authenticated source. Thus,

the inconsistency in LARP1B expression could be a result of genetic instability and

phenotypic drift between different cell line batches. Many cancer cell lines have

defects in genes that monitor and repair DNA damage, giving rise to an increased

mutation rate. Therefore, the genotype of continuous cell lines can change with time

and is likely to progress the longer the cell line is cultured [265]. Genetic instability

in a cell line can also be influenced by the confluence of the cells at the time of

subculture and their maintenance in the correct growth media. As LARP1B was only

one gene out of eight whose expression was inconsistent with previous data, we

decided to include it in the siRNA-mediated knock-down screen.

As expected, reovirus infection caused very little cytotoxicity in an untransformed

normal human lung fibroblast cell line (MRC-5) and PBMCs isolated from the blood

of a healthy human donor, compared to the SCCHN cell lines. Cell lines derived from

normal human tissue are extremely difficult to grow and maintain, and therefore a

normal head and neck cell line was not available to us for direct comparison to the

SCCHN cell lines. It could be argued that both MRC-5 and PBMCs have limitations

as they do not originate from the head and neck region and do not contain any

epithelial features. However, our findings still demonstrate the selective nature of

reovirus as an oncolytic agent, just as a plethora of research has documented since the

1970s [124]. Further strengthening this point, many studies have used both human

and mouse fibroblast cell lines (most notably NIH-3T3) for similar purposes [125,

133, 134, 142]. PBMCs have been shown to transport and protect reovirus particles

from neutralising-anti-reovirus antibodies after IV injection in cancer patients [179,

225], but are not known to be susceptible to reovirus oncolysis, which is consistent

with our data.

104

Having successfully optimised the conditions for siRNA-mediated transfection in the

PJ41 cell line using the KDalert™ GAPDH assay kit, we transiently transfected PJ41

cells with 2 specific siRNAs for SLCO1B3, MGMT, SLC36A4, YAP1, ZNF600,

P2RY6, BIRC2 and LARP1B. The reduction in mRNA expression of all genes proved

to be very efficient and ranged from 79.1 to 99.9% knock-down compared to negative

siRNA control treated cells. Knock-down of SLCO1B3, MGMT, SLC36A4, ZNF600,

P2RY6, BIRC2 and LARP1B had little effect on reovirus-mediated cell death.

However, most interestingly, siRNA-mediated knock-down of YAP1 caused

significant sensitisation to reovirus treatment at all MOIs tested (apart from MOI

1000). At MOI 1000, a high percentage of cells had succumbed to death, and this

could explain why no major sensitization was seen when YAP1 was reduced. The

reovirus IC50 value of the PJ41 cells with YAP1 knock-down was MOI 115.9, whilst

the IC50 for the negative control siRNA-treated cells was MOI 365.7. Therefore

knock-down of YAP1 caused approximately a 3-fold increase in the sensitivity of the

cells to reovirus, although still not to the level of the HN5 and PJ34 cell lines. There

was no major differences in OD values between un-infected cells treated with

negative control siRNA and un-infected cells treated with the target-gene siRNA.

This implied that the reovirus sensitivity was indeed due to reduced YAP1, and was

not simply a result of differences in transfection-associated cytotoxicity.

siRNA-mediated knock-down of the YAP1 protein proved to be less efficient than

knock-down of YAP1 at the mRNA level. It is well-known that levels of RNA and

their protein products can vary considerably, as there are many processes that occur

between transcription and translation, including RNA processing, alternative and

differential splicing, and protein modifications. Efforts were made to improve this

reduction by altering the transfection conditions such as the cell seeding density, the

concentration of the transfection agent and the concentration of the siRNAs, but this

did not enhance the knock-down, or compromised the cellular viability. Although the

reduction in YAP1 was substantial enough to influence reovirus-induced cell death,

perhaps an even greater sensitisation would have been observed if YAP1 had been

completely eliminated. We now know that there are techniques that are capable of

permanently modifying genomic DNA, such as Clustered Regularly Interspaced Short

Palindromic Repeats (CRISPR). However, this is a relatively new system that was

not available for commercial use when this research was conducted. Our results

105

suggest that YAP1 may contribute to reovirus-resistance in SCCHN, but probably

functions in combination with other proteins to supress reovirus-induced cell death.

An extensive search of the literature revealed no evidence linking YAP1 to an anti-

viral intracellular host response pathway, but this research suggests that it could be

part of one. Oka et al found that YAP2, an alternative YAP isoform, forms

complexes with zona-occluden-2 (ZO-2) at tight junctions via their PDZ-binding

domains [266]. The main reovirus cellular receptor, junction-adhesion molecule-A

(JAM-A), has been shown to associate with several different PDZ-domain containing

proteins, including ZO-2 [267, 268]. Thus, one possible theory is that YAP1

indirectly prevents viral entry to the cell through ZO-2 interaction with JAM-A at the

cell-surface membrane. These are hypothetical mechanisms that may support YAP1-

associated reovirus-resistance in SCCHN. However, more evidence is required to

claim that YAP1 is an important factor in more SCCHN cell lines, which will be

explored in later chapters.

YAP1 is a major downstream target of the Hippo signalling pathway. This pathway

becomes activated to inhibit cell proliferation by preventing the nuclear localisation

and activation of YAP1 [250, 252], as detailed in Chapter 4. Intriguingly, there is

evidence suggesting that two out of the eight genes tested in this project are

interlinked. BIRC2 and YAP1 are located in close proximity on chromosome 11q, and

they can act independently as oncogenes or can synergise to promote tumorigenesis

by virtue of their co-amplification at the same genomic locus [255]. Elevated

expression of cIAP1 (the protein product of BIRC2) and YAP mRNA and protein was

found in human and mouse hepatocellular carcinoma tumours that contained an

amplicon in the 11q22 region [255]. Coincidentally, cIAP1 can also be degraded to

allow reovirus-induced apoptosis to proceed in some infected cells [242, 243]. Our

results showed that knock-down of BIRC2 in the PJ41 cell line had no significant

effect on reovirus-induced cell death, suggesting that YAP1 promotes resistance to

reovirus independently of BIRC2. It would however, be of value to analyse the effect

of the simultaneous knock-down of YAP1 and BIRC2 in PJ41 cells after infection with

reovirus, and whether the expression level of one gene affects the other. The YAP1

gene locus is also often amplified in other human cancers, including oral squamous

cell carcinoma tissues [269, 270]. Impairment of Hippo signalling and nuclear

location of YAP often results in tumorigenesis. Increased YAP1 protein expression

106

has been observed in ovarian, colon, lung, breast and prostate cancers [219, 254, 255,

271]. The expression and localisation of YAP1 in head and neck carcinoma tissues in

comparison to normal tissues will be explored in Chapter 5.

107

3.5. CONCLUSION

PJ34, HN5 and PJ41 SCCHN cell lines displayed low, medium and high resistance to

reovirus oncolysis respectively. The same 3 cell lines in turn showed, low, medium

and high mRNA expression of SLCO1B3, MGMT, SLC36A4, YAP1, ZNF600, P2RY6,

BIRC2, but not LARP1B. Our findings generally agreed with earlier published data

and preliminary work performed elsewhere. This justified testing the potential

influence that these genes may have in predicting the susceptibility of SCCHN cell

lines to reovirus oncolysis. siRNA-mediated knock-down of SLCO1B3, MGMT,

SLC36A4, ZNF600, P2RY6, BIRC2 and LARP1B had little effect on reovirus-

mediated cell death in the PJ41 cell line. However, knock-down of YAP1 caused

significant sensitisation to reovirus treatment, suggesting that a certain level of YAP1

expression in the cell may contribute to reovirus resistance in SCCHN. Further

experiments in the other SCCHN cell lines are needed to draw more comprehensive

conclusions, which will be explored in later chapters.

108

CHAPTER 4

TARGETING YES-ASSOCIATED PROTEIN-1

(YAP1) AS A FACTOR THAT INFLUENCES

REOVIRUS ONCOLYSIS IN SCCHN CELL LINES

109

4. TARGETING YES-ASSOCIATED PROTEIN-1 (YAP1) AS A FACTOR THAT

INFLUENCES REOVIRUS ONCOLYSIS IN SCCHN CELL LINES

4.1. INTRODUCTION

Understanding the mechanism of reovirus-induced cancer cell death could lead to the

discovery of new biomarkers of treatment response in patients receiving reovirus

therapy. Such biomarkers would be particularly beneficial to squamous cell

carcinoma of the head and neck (SCCHN) patients, as Reolysin® has reached Phase

III clinical testing status in this cancer type. Although the initial results of this trial

looked encouraging, it is not certain whether the primary endpoints of evaluating

overall survival and progression-free survival after administration with Reolysin®,

have been met. Biomarkers of treatment response may help improve clinical trial

design and outcome. The previous chapter demonstrated that siRNA-mediated knock-

down of yes-associated protein-1 (YAP1) in the PJ41 SCCHN cell line caused

significant sensitisation to reovirus oncolysis. Further experiments were needed to

establish whether YAP1 influences the other SCCHN cell lines to reovirus treatment,

which is the purpose of this chapter.

The Hippo signalling pathway is an evolutionary conserved regulator of cell

proliferation and apoptosis. The components of the pathway were originally

identified in Drosophila melanogaster using genetic screens that were devised to

discover novel tumour suppressor genes [272]. The pathway is conserved in

vertebrates, including mammals. The Hippo pathway is activated to prevent cell

proliferation when cells become too confluent, or by cell stress to induce apoptosis

[273]. The core of the mammalian pathway is composed of a pair of serine/threonine

kinases, mammalian STE20-like protein kinase-1 and -2 (MST1 and MST2), and

large tumour suppressor-1 and -2 (LATS1 and LATS2). The core also comprises the

adaptor proteins Salvador homologue 1 (SAV1), and MOB kinase activator 1A and

1B (MOB1A and MOB1B) [251]. When the Hippo pathway becomes stimulated,

MST1/2 kinases become activated, which phosphorylate and subsequently activate

other members of the complex [272]. LATS1/2 directly phosphorylate downstream

YAP and its co-protein, transcriptional co-activator with PDZ-binding motif (TAZ).

Phosphorylation of YAP on serine 127 (S127) and TAZ at serine 89 (S89) serve as

110

docking sites for 14-3-3 proteins, and this interaction leads to the cytoplasmic

retention of YAP and TAZ where they remain inactive. Mutation of the S127 residue

has been shown to activate YAP [252], highlighting the importance of this residue in

Hippo signalling. The phosphorylation of another serine residue, serine 381 (S381) of

YAP1 and S311 of TAZ, leads to the poly-ubiquitination and degradation of YAP and

TAZ [274]. In the absence of Hippo signalling, YAP/TAZ migrate to the nucleus.

YAP/TAZ do not directly bind to DNA in the nucleus, but are cofactors that regulate

gene expression by interacting with transcription factors, including the TEA domain-

containing transcription factor family (TEAD). This interaction stimulates expression

of genes such as c-myc, SOX4, AFP, MK167, AREG, CTGF and CCND1, which

promote proliferation and inhibit apoptosis [219, 253, 275]. Figure 4.1 illustrates the

known proteins involved in Hippo pathway signalling.

There are multiple upstream signals that are known to activate the Hippo pathway. At

cell-cell junctions, the Merlin protein (encoded by the NF2 gene), has been shown to

aid the assembly of protein scaffolds such as Kibra, that allows the activation of the

LATS kinases and the phosphorylation of YAP [276]. The Crumbs complex (CRB) is

a polarity protein that can bind to and sequester YAP/TAZ to the cytoplasm, as can

the angiomotin (AMOT) protein [276]. Complexes of -catenin at E-cadherin

junctions can inhibit YAP nuclear accumulation [276]. Apicobasal cell polarity

(ABCP) proteins, such as mammalian Scribble (SCRIB) can serve as an adaptor to

facilitate activation of the core kinase cassette [251]. Delocalisation of SCRIB from

the plasma membrane is common in cancer, and is associated with epithelial-

mesenchymal transition (EMT) and YAP/TAZ nuclear activation [276]. Yu et al

showed that G-Protein coupled receptors (GPCRs) can positively or negatively

regulate the Hippo pathway. Activation of Gs-coupled receptors increases LATS1/2

kinase activity, whereas activation of G12/13 or Gq/11-coupled receptors inhibits

LATS1/2 kinases, resulting in YAP activation. Rho GTPases and the actin

cytoskeleton are located between these GPCRs and the LATS kinases, and appear to

be involved in this process [220].

There are at least eight known isoforms of the YAP protein that are generated by

differential splicing. The two major isoforms are YAP1 and YAP2, which differ by

the presence of one or two WW domains respectively (Figure 4.2). At the amino

111

terminus, there is a TEAD factor binding domain that contains the S127 residue [277].

Complexes between the YAP WW domain(s) and the PPxY motif-containing LATS1

kinase are important in inhibiting the proliferative activity of YAP by preventing its

localisation to the nucleus [277]. YAP also contains a SH3 binding motif, a

transcriptional activation domain (TAD) and a PDZ-binding motif. The PDZ-binding

motif is composed of amino acids that are essential for the nuclear translocation of

YAP [277].

Figure 4.1. A schematic representation of the proteins involved in the mammalian Hippo

pathway. Supposed tumour suppressors are shown in blue and supposed oncogenes are shown in red.

Imaged adapted from [251]. Compared to other well-defined pathways, Hippo signalling is a relatively

new concept and more components of this pathway have yet to be characterised [278].

112

Figure 4.2. Functional domains of YAP1 and YAP2; the two major isoforms of the YAP protein

[277].

In human cancers, germline or somatic mutations in Hippo pathway genes are rare.

An exception to this is the inherited mutation of the NF2 gene, which causes an

autosomal dominant syndrome called type 2 neurofibromatosis, giving rise to tumours

of the brain and spinal cord [251]. Although mutations in Hippo pathway genes are

not common in human cancers, convincing evidence supports a role of this pathway in

human tumourigenesis. Down-regulation of upstream kinases LATS1/2 and MST1/2

has been reported in various cancer types in humans, including soft tissue sarcomas

[279], retinoblastomas [280] and acute lymphoblastic leukemia [281]. Studies in mice

have implicated MST1/2 as tumour suppressors. Zhou et al demonstrated that

MST1/2 deficiency in the liver results in loss of inhibitory phosphorylation at S127 of

YAP1, huge overgrowth, and hepatocellular carcinoma (HCC) [282]. The nuclear

location of YAP/TAZ often results in tissue overgrowth and tumourigenesis. In

normal human tissues, YAP is reported to be infrequently nuclear [251, 254].

Deregulation of the Hippo pathway is commonly associated with poor patient

prognosis [251, 271, 283].

Interestingly, there is evidence linking the interaction of oncogenic viruses to Hippo

signalling [284], but information of this pathway being modulated by oncolytic

viruses is scarce. For example, Hepatitis B virus (HBV) disturbs the normal control

113

of Hippo signalling through up-regulation of YAP [285]. Human papillomavirus

(HPV) and Epstein-Barr virus (EBV) both down-regulate E-cadherin expression and

promote WNT signalling, which is closely inter-linked with the Hippo pathway [286-

289]. Human T lymphotropic virus type 1 (HTLV-1) perturbs the expression of

complexes such as SCRIB that are central to the regulation of cell polarity, and

modulates WNT signalling [290]. The Hippo pathway has been shown to mediate the

oncogenic activity of Kaposi-sarcoma-associated herpesvirus (KSHV). KSHV

encodes a viral G-protein-coupled receptor (vGPCR) that acts through the G-proteins,

Gq/11 and G12/13, to inhibit the Hippo pathway kinases LATS1/2, thus promoting the

activation and nuclear accumulation of YAP and TAZ to initiate the progression of

the AIDS-defining cancer, Kaposi sarcoma (KS) [291]. The authors of this paper

suggest using inhibitors of YAP for the prevention of KS. Our results so far imply

that high expression of YAP1 contributes to resistance to oncolytic reovirus in

SCCHN. In this context, inhibitors of YAP might increase the likelihood of an

effective anti-cancer response to reovirus infection in SCCHN.

114

4.2. STUDY OBJECTIVE

The objective of this chapter was study to the effect of plasmid-mediated over-

expression of YAP1 in SCCHN cell lines that exhibited low endogenous expression

of YAP1 and a high level of sensitivity to reovirus oncolysis.


1. Transient over-expression of YAP1 in the PJ34 SCCHN cell line and assessment

of the resultant effect on cell survival after infection with reovirus by the MTS

assay.

2. Stable over-expression of YAP1 in the HN5 SCCHN cell line and determination

of the resultant effect on cell survival after infection with reovirus by the MTS

assay.

3. Transient over-expression of YAP1 in the COS-1 monkey fibroblast cell line and

assessment of the resultant effect on cell survival after infection with reovirus by

the MTS assay, to assess whether the effect is SCCHN cell line specific.

4. Immunofluorescent staining and confocal imaging in SCCHN cell lines to

determine the cellular localisation and possible function of the YAP1 protein,

which may be an important factor in how it mediates reovirus oncolysis.

5. Treatment of the PJ41 SCCHN cell line with sphingosine-1-phosphate (S1P) to

stimulate the de-phosphorylation and nuclear expression of YAP1, and the

subsequent determination of the effect on cell survival after infection with

reovirus by the MTS assay.

115

4.3. RESULTS

4.3.1. Transient over-expression of YAP1 in the PJ34 SCCHN cell line

In order to over-express the YAP1 gene in the PJ34 cell line, which displayed the

lowest expression of YAP1 and was the most sensitive to reovirus oncolysis, a

transient transfection with 2 YAP1-containing plasmids was performed (Section

2.16.1). An EYFP-tagged-YAP1 plasmid and a Flag-tagged-YAP1 plasmid was

provided to us by Dr Nic Tapon, Cancer Research UK London Research Institute.

Cells were also treated with lipofectamine transfection agent alone, or with media

alone (un-transfected sample), which served as controls. At 24 hours post-

transfection, RNA was extracted from the cells and the cDNA template was used to

quantify YAP1 mRNA expression by RT-qPCR (Sections 2.10, 2.11 and 2.12). The

relative expression value was calculated as a ratio to the housekeeping gene β-actin.

The transfection conditions were initially optimised by using 3 different

concentrations of lipofectamine (0.25, 0.35 and 0.45µL/well). Figure 4.3 shows that

YAP1 expression was enhanced when higher concentrations of lipofectamine was

used to deliver the plasmids. The relative mRNA expression of YAP1 in cells

transfected with the EYFP-YAP1 and Flag-YAP1 plasmids was 91-fold and 94-fold

higher than the lipofectamine control respectively, when 0.45µL/well was used.

PJ34 cell lysates were also analysed for YAP1 protein expression by western blotting

(Section 2.14). There was considerably more YAP1 protein in the cells transfected

with both YAP1 plasmids compared to cells treated with lipofectamine and media

alone. A band at 81kDa was detected in the cells transfected with the EYFP-YAP1

plasmid, which corresponded to the expected combined molecular weight of the

YAP1 protein (54kDa) and the EYFP-tag (27kDa). Similarly, a band at 56kDa was

detected in the cells treated with the Flag-YAP1 plasmid, which related to the

combined molecular weights of YAP1 (54kDa) and the Flag-tag (2kDa). Endogenous

YAP1 was detected in the samples at 54kDa. Again, there was a lipofectamine-dose-

response increase in YAP1, with 0.45µL/well showing the greatest YAP1 protein

over-expression (Figure 4.4 A). The combined intensity of the endogenous and

exogenous YAP1 bands in each sample was quantified and normalised to their

corresponding β-actin bands by densitometry analysis (Section 2.14.4) (Figure 4.4

116

B). This revealed a 9-fold and 16-fold increase in total YAP1 protein expression in

cells transfected with EYFP-YAP1 and Flag-YAP1 plasmids respectively, compared to

cells treated with 0.45µL/well lipofectamine alone. Thus, 0.45µL/well was chosen as

the optimal concentration of lipofectamine to deliver the plasmids.

Figure 4.3. YAP1 mRNA expression in the PJ34 SCCHN cell line after transient over-expression

of YAP1. cDNA from PJ34 cells treated with the Flag-YAP1 plasmid (blue bars), EYFP-YAP1 plasmid

(red bars), and lipofectamine or media alone (green bars), was analysed by RT-qPCR. The mRNA

expression of YAP1 is shown relative to the housekeeping gene β-actin (×1000). The highest

concentration of lipofectamine (0.45µL/well) caused the greatest increase in YAP1 expression. Cells

transfected with the EYFP-YAP1 and Flag-YAP1 plasmids generated a 91-fold and 94-fold increase in

YAP1 expression compared to the lipofectamine control respectively, when 0.45µL/well was used. Error bars represent the SD from triplicate samples.

0

50

100

150

200

250

300

350

0.4

5µ

L/w

ell

Lip

o

0.3

5µ

L/w

ell

Lip

o

0.2

5µ

L/w

ell

Lip

o

0.4

5µ

L/w

ell

Lip

o

0.3

5µ

L/w

ell

Lip

o

0.2

5µ

L/w

ell

Lip

o

0.4

5µ

L/w

ell

Lip

o

0.3

5µ

L/w

ell

Lip

o

0.2

5µ

L/w

ell

Lip

o

Flag-YAP1 EYFP-YAP1 Lipofectamine only Media

only

Rel

ati

ve

mR

NA

ex

pre

ssio

n

117

A

B

Figure 4.4. YAP1 protein expression in the PJ34 cell line after transient over-expression of

YAP1. Whole cell lysates were collected from cells treated with the Flag-YAP1 plasmid, EYFP-YAP1

plasmid, lipofectamine alone and media alone. A. YAP1 protein expression was determined by

western blotting. Endogenous YAP1 was detected in the samples at 54kDa. EYFP-tagged-YAP1 was

detected at 81kDa, whereas Flag-tagged-YAP1 was detected at 56kDa. There was a clear over-

expression of YAP1 in the plasmid-transfected cells compared to the cells treated with lipofectamine or

media alone. B. Total YAP1 protein expression (endogenous and exogenous combined) in each

sample was normalised to the -actin loading control and quantified by densitometry using LI-COR

Image Studio Lite 4.0.21 software. This confirmed that transfection with EYFP-YAP1 (red bars) and

Flag-YAP1 (blue bars) plasmids caused a 9-fold and 16-fold increase in YAP1 expression respectively,

compared to cells treated with 0.45µL/well lipofectamine alone (green bars).

0

10

20

30

40

50

60

70

0.4

5µ

L/w

ell

Lip

o

0.3

5µ

L/w

ell

Lip

o

0.2

5µ

L/w

ell

Lip

o

0.4

5µ

L/w

ell

Lip

o

0.3

5µ

L/w

ell

Lip

o

0.2

5µ

L/w

ell

Lip

o

0.4

5µ

L/w

ell

Lip

o

0.3

5µ

L/w

ell

Lip

o

0.2

5µ

L/w

ell

Lip

o

Flag-YAP1 EYFP-YAP1 Lipofectamine only Media

only

Rel

ati

ve

den

sity

118

4.3.2. Transient YAP1 over-expression caused increased resistance of the PJ34

cell line to reovirus-induced cell death

Having successfully achieved transient YAP1 over-expression in the PJ34 cell line, the

susceptibility of these cells to reovirus infection was evaluated, in comparison to cells

treated with lipofectamine or media alone (Section 2.16.3). First, the average OD of

each un-infected transfection condition was compared to check for cytotoxic effects.

After 24 hours post-transfection, cells were treated with media alone for a further 48

hours before analysis via the MTS assay (Section 2.9). There was no cytotoxicity

observed in the cells treated with 0.45µL/well lipofectamine alone. However, there

was some cytotoxicity observed in the cells transfected with the DNA plasmids

(Figure 4.5).

In order to account for the differences in transfection-associated cytotoxicity, after 24

hours post-transfection, cells from each treatment condition were counted prior to

infection with reovirus. The cell counts were used to calculate the required MOI per

well (Section 2.8), which ensured that the same number of virus particles per cell was

used in each treatment condition. Cells were infected with serial dilutions of reovirus

starting at MOI 1000. At 48 hours post-infection with reovirus, the % cell survival in

each treatment condition was assessed by the MTS assay (Section 2.9) and the IC50

values were determined using CalcuSyn software (Biosoft, UK) (Section 2.27.4)

Figure 4.6 shows that plasmid-mediated over-expression of YAP1 in the PJ34 cell line

caused a significant increase in resistance to reovirus oncolysis at all MOIs tested

(p<0.05 by un-paired t-test). Compared to cells treated with lipofectamine or media

alone (both had reovirus IC50 values of MOI 6.7), there was a 12-fold and 4-fold

increase in resistance in cells transfected with EYFP-YAP1 (IC50 MOI 77.6) and Flag-

YAP1 (IC50 MOI 27.9) plasmids respectively. This suggested that over-expression of

YAP1 may be a factor that promotes reovirus resistance in the PJ34 SCCHN cell line.

119

Figure 4.5. The transfection-associated toxicity in the PJ34 cell line. After 24 hours transient-

transfection, PJ41 cells were treated with media alone for a further 48 hours before analysis via the

MTS assay. The raw Optical Density (OD) values for each treatment was compared. There was some

cytotoxicity observed in the cells transfected with EYFP-YAP1 and Flag-YAP1 plasmids. The graph

shows the mean OD of 2 assay repeats and error bars represent SD.

Med

ia o

nly

Lip

ofe

cta

min

e o

nly

EY

FP

-YA

P1

Fla

g-Y

AP

1

0 .0

0 .5

1 .0

1 .5

T ra n s fe c tio n c o n d it io n

Op

tic

al

De

ns

ity

(4

90

nm

)

120

Figure 4.6. Evaluation of reovirus-induced cell death after plasmid-mediated over-expression of

YAP1 in the PJ34 cell line. Cells were transiently transfected with A. the EYFP-YAP1 plasmid, or B.

the Flag-YAP1 plasmid (red circles). Cells were also treated with lipofectamine only (green triangles)

or media only (blue triangles). After counting the cells in each treatment condition, cells were

subsequently infected with serial dilutions of reovirus, starting at MOI 1000. The % cell survival in

each treatment condition was then assessed using the MTS assay. The IC50 values of each treatment

condition were determined using CalcuSyn software and are shown in a table below each graph. Over-

expression of YAP1 enhanced the resistance to reovirus-induced cell death at all MOIs tested. *p<0.05,

**p<0.01, ***p<0.001 and ****p<0.0001 by un-paired t-test, with respect to the cells treated with

lipofectamine alone. Error bars represent the SD from 2 assay repeats.

0.0

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

E Y F P -Y A P 1

L ip o fe c ta m in e o n ly

M e d ia o n ly

***

**

** *

**** ****


EYFP-YAP1 77.6 ± 7.8

Lipofectamine only 6.7 ± 0.8

Media only 6.7 ± 0.4

0.0

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

F la g -Y A P 1


M e d ia o n ly

**

**

**

***

******


FLAG-YAP1 27.9 ± 3.1



A

B

121

4.3.3. Stable over-expression of YAP1 in the HN5 cell line

Next, we intended to create stable transfected cell lines that over-expressed YAP1

long-term. Although transiently transfected cells express the foreign gene, they do

not integrate it into their genomes. Because the new gene will not be replicated, the

cells can only express the gene for a finite period of time. Conversely, stable

transfected cell lines incorporate the foreign gene into their genome, which means that

descendants of these transfected cells will also express the new gene [292]. The

plasmids used in this study contained a neomycin antibiotic resistance gene, which

allowed the selection of cell colonies with the G418 antibiotic. Therefore cells that

had successfully incorporated the plasmid into their genomes were able to survive,

whereas cells that failed to uptake the plasmid were killed by the antibiotic. Several

attempts were made to create a stable cell line that over-expressed YAP1 in PJ34 cells,

but none of the clones survived long enough in culture to test their reovirus

susceptibilities, even after modifying the transfection parameters. We attempted to

transfect PJ34 cells with three other YAP1-containing plasmids from Origene

Technologies (Rockville, USA). Stable colonies using these plasmids did survive in

culture, but compared to PJ34 parental cells, none of them significantly over-

expressed YAP1 at the mRNA or protein levels. Thus, stable clones were created

using the HN5 cell line (Section 2.16.2), which was the second most sensitive line to

reovirus oncolysis and expressed lower endogenous levels of YAP1 than the PJ41 cell

line. Stable clones were created after transfection with the Flag-YAP1 plasmid or an

empty vector (EV) control plasmid. The EV plasmid was a pcDNA3.1 vector, which

contained the same components as the Flag-YAP1 plasmid but without the YAP1-tag

insert, and was provided to us by Dr Lisi Meira (The University of Surrey).

Therefore, the EV plasmid served as a non-specific negative control to the Flag-YAP1

plasmid. In addition, stable clones were generated after transfection with the EYFP-

YAP1 plasmid (Section 2.16.2). The EYFP-YAP1 plasmid was a different vector type

(i.e. not a pcDNA3.1 vector), and an EV control was not available for direct

comparison. The HN5 parental cell line was therefore used as a negative control for

the EYFP-YAP1 plasmid.

13 EYFP-YAP1 clones were selected and the resultant YAP1 protein expression levels

were compared to the HN5 parental cell line by western blotting (Section 2.14). 9 of

122

the clones produced an EYFP-YAP1 exogenous band at 81kDa, as well as a band at

54kDa, which signified endogenous YAP1 expression (Figure 4.7A). The combined

intensity of the exogenous and endogenous YAP1 bands in each sample was

quantified and normalised to their corresponding β-actin bands by densitometry

analysis (Section 2.14.4) (Figure 4.7 B). EYFP-YAP1-clone 6 produced the greatest

increase in total YAP1 protein expression, and was 5-fold higher than total YAP1 in

the HN5 parental cell line. Out of the 5 Flag-YAP1 clones tested, clone 2 produced

the most intense band at 56kDa, and there was a 4-fold increase in total YAP1

compared to the HN5 parental cell line, as confirmed by densitometry analysis

(Figure 4.8 A and B). As expected, there was no change in total YAP1 protein

expression in the EV-clones compared to the HN5 parental cell line (Figure 4.8 A

and B).

123

A

B

Figure 4.7. YAP1 protein expression in HN5 SCCHN cell line clones after stable over-expression

of YAP1 using the EYFP-YAP1 plasmid. Whole cell lysates were collected from 13 EYFP-YAP1

stable clones and from the HN5 parental cell line. A. YAP1 protein expression was determined by

western blotting. Endogenous YAP1 was detected in the samples at 54kDa, whereas exogenous YAP1

was detected at 81kDa. The 42kDa -actin bands in each sample generally showed uniform protein

loading. B. Total YAP1 expression of each band was quantified and normalised to the corresponding

-actin loading control bands by densitometry analysis. This confirmed that out of all the clones tested

(purple bars), clone 6 expressed the highest levels of YAP1 compared to the HN5 parental cell line

(black bar), as shown by the red arrow.

0

20

40

60

80

100

120

140

clo

ne

2

clo

ne

3

clo

ne

4

clo

ne

5

clo

ne

6

clo

ne

7

clo

ne

8

clo

ne

9

clo

ne

10

clo

ne

11

clo

ne

12

clo

ne

13

clo

ne

14

HN

5 p

are

nta

l

EYFP-YAP1

Re

lati

ve d

en

sity

124

A

B

Figure 4.8. YAP1 protein expression in HN5 SCCHN cell line clones after stable over-expression

of YAP1 using the Flag-YAP1 plasmid. Whole cell lysates were collected from 5 Flag-YAP1 stable

clones, 3 empty vector (EV) stable clones, and from the HN5 parental cell line. A. YAP1 protein

expression was determined by western blotting. Endogenous YAP1 was detected in the samples at

54kDa, whereas exogenous YAP1 was detected at 56kDa. The 42kDa -actin bands in each sample

showed uniform protein loading. B. Total YAP1 expression of each band was quantified and

normalised to the corresponding -actin loading control bands by densitometry analysis. This

confirmed that out of the 5 Flag-YAP1 clones tested (green bars), Flag-YAP1 clone 2 expressed the

highest levels of YAP1 compared to the HN5 parental cell line (black bar). There was no alteration in

YAP1 protein expression in the EV clones (blue bars). Flag-YAP1 clone 2 and EV-clone 1 were

chosen for further analysis, as shown by the red arrows.

125

4.3.4. Stable over-expression of YAP1 caused increased resistance to reovirus-

mediated cell death in the HN5 cell line

Having achieved stable over-expression of the YAP1 protein in the HN5 cell line, the

next step was to determine the susceptibility of the stable clones to reovirus oncolysis

(Section 2.16.3). Unlike transient transfection, stable transfection of a gene of

interest persists after several cellular passages. Therefore, this allowed us to study the

effect of YAP1 over-expression at earlier and later times of reovirus infection. EYFP-

YAP1-clone 6 and Flag-YAP1-clone 2 were selected because they exhibited the

highest protein levels of YAP1. EV-clone 1 was also selected for use as a non-

specific control to the Flag-YAP1-clone 2.

Prior to determining the susceptibility of the stable clones to reovirus-induced cell

death, we evaluated the proliferation rate of each clone compared to the HN5 parental

cell line. The average OD was compared at 48, 72 and 96 hours after seeding the cells

in tissue culture plates, via analysis by the MTS assay (Section 2.9). Figure 4.9

demonstrates that stable over-expression of YAP1 using both YAP1-plasmids caused

an increase in cellular proliferation over time, compared to the EV-clone 1 and HN5

parental cell line.

In order to account for the differences in proliferation rate, HN5 parental cells, and

cells from each stable clone, were counted prior to infection with reovirus. The cell

counts were used to calculate the required MOI per well (Section 2.8), which ensured

that the same number of virus particles per cell was used for each clone or cell line.

Cells were infected with serial dilutions of reovirus starting at MOI 500 or 1000. At

24, 48, and 72 hours post-infection with reovirus, the % cell survival in each treatment

condition was assessed by the MTS assay (Section 2.9) and the IC50 values were

determined using CalcuSyn software (Section 2.27.4).

Figure 4.10 shows that the stable EYFP-YAP1-clone 6 (which exhibited a 5-fold

over-expression of YAP1 compared to HN5 parental cells) caused a significant

increase in resistance to reovirus oncolysis at all time-points and at all MOIs tested

(p<0.05 by un-paired t-test), compared to the HN5 parental cell line. The reovirus

IC50 of EYFP-YAP1-clone 6 was 8-fold, 6-fold and 3-fold more resistant than the

HN5 parental cell line at 24, 48 and 72 hours post infection respectively.

126

The reovirus sensitivity was also evaluated between the Flag-YAP1-clone 2 and EV-

clone 1 stable cell lines (Figure 4.11). Compared to the HN5 parental cell line, there

was some non-specific reovirus resistance observed in the EV-clone 1 control.

However, this non-specific effect was less apparent at later times of infection (48 and

72 hours post infection). To check that the non-specific effect caused by the EV

plasmid was not due to an erroneous gene insert, the plasmid was sent for DNA

sequencing analysis (Sanger sequencing facility, Department of Biochemistry,

University of Cambridge). Results showed that the EV plasmid contained no extra

elements, which confirmed its use as a valid negative control. Both the EYFP-YAP1

and Flag-YAP1 plasmids were also sent for analysis and the results verified that both

vectors contained the correct YAP1 nucleotide sequence. Therefore, in order to assess

the specific effect of YAP1 over-expression, all statistical comparisons were made

directly to the EV-clone 1 stable cell line. Compared to EV-clone 1, Flag-YAP1 clone

6 (which displayed a 4-fold over-expression of YAP1 compared to HN5 parental

cells) was 2-fold more resistant to reovirus at 24 hours, and 3-fold more resistant at 48

and 72 hours post-infection. This was statistically significant at the majority of MOIs

tested, apart from at the very high MOIs, which expectedly caused considerable cell

death. Taking into account both the transient over-expression of YAP1 in PJ34 and

the stable over-expression of YAP1 in HN5, these results implied that a certain level

of YAP1 expression in SCCHN cells may contribute to reovirus resistance.

127

A

B

Figure 4.9. The differences in proliferation rates between the stable clones and the HN5 parental

cell line. The average Optical Density (OD) values were compared over a time-course of 48 hours

(pink bars), 72 hours (green bars) and 96 hours (blue bars) in culture between A. the EYFP-YAP1-clone

6 and the HN5 parental cell line, and B. the Flag-YAP1-clone 2, the empty vector (EV)-clone 1 and the

HN5 parental cell line. OD values were determined by analysis via the MTS assay. The clones that

over-expressed YAP1 proliferated more quickly over time than the HN5 parental cell line and the EV-

clone 1 control. The graph shows the mean OD of 2 assay repeats and error bars represent SD.

Fla

g-Y

AP

1-c

lon

e 2

EV

-clo

ne 1

HN

5 p

are

nta

l

0 .0

0 .5

1 .0

1 .5

Op

tic

al

De

ns

ity

(4

90

nm

)

4 8 h o u rs

7 2 h o u rs

9 6 h o u rs

EY

FP

-YA

P1

-clo

ne 6

HN

5 p

are

nta

l

0 .0

0 .5

1 .0

1 .5

Op

tic

al

De

ns

ity

(4

90

nm

)

4 8 h o u rs

7 2 h o u rs

9 6 h o u rs

128

Figure 4.10. Stable over-expression of the YAP1 protein using the EYFP-YAP1 plasmid, promoted resistance to reovirus in the HN5 SCCHN cell line. Stable clones were generated from the HN5 cell line by transfection with the EYFP-YAP1 plasmid. The HN5 parental cell line (blue triangles) and EYFP-YAP1-

clone 6 (which displayed a 5-fold over-expression of YAP1 compared to wild-type HN5) (red circles) were infected with serial dilutions of reovirus starting at MOI

500 for A. 24, B. 48 or C. 72 hours. The % cell survival was then assessed using the MTS assay. The IC50 values of the cell lines were determined using CalcuSyn

software and are shown in a table below each graph. Stable over-expression of YAP1 enhanced the resistance to reovirus-induced cell death at all MOIs tested

compared to the HN5 parental cell line. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001 by un-paired t-test. Error bars represent the SD from 2 assay repeats.

0.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

0

2 5

5 0

7 5

1 0 0

2 4 h o u r s


% c

ell

su

rv

iva

l H N 5 p a re n ta l

E Y F P -Y A P 1 c lo n e 6

*****

***

******

****

****

****

A


EYFP-YAP1-clone 6 187.5 ± 9.4

HN5 parental 25.0 ± 0.8

0.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

0

2 5

5 0

7 5

1 0 0

4 8 h o u r s


% c

ell

su

rv

iva



***

*****

***

***

***

****

****

B




0.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

0

2 5

5 0

7 5

1 0 0

7 2 h o u r s


% c

ell

su

rv

iva



* ** **

**

***

****

**

C




129

Figure 4.11. Stable over-expression of the YAP1 protein using the Flag-YAP1 plasmid, promoted resistance to reovirus in the HN5 SCCHN cell line. Stable

clones were generated from the HN5 cell line by transfection with the Flag-YAP1 plasmid or the empty vector (EV)-control plasmid. The HN5 parental cell line

(blue triangles), EV-clone1 (green triangles), and Flag-YAP1-clone 2 (which displayed a 4-fold over-expression of YAP1 compared to wild-type HN5) (red circles)

were infected with serial dilutions of reovirus starting at MOI 1000 for A. 24, B. 48 or C. 72 hours. The % cell survival was then assessed using the MTS assay.

The IC50 values of the cell lines were determined using CalcuSyn software and are shown in a table below each graph. There was some non-specific resistance

caused by the EV-clone 1, although this did decline at later times of infection. Compared to EV-clone 1, stable over-expression of YAP1 using Flag-YAP1-clone 2

enhanced the resistance to reovirus at most MOIs tested (*p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001 by un-paired t-test). Error bars represent the SD from 2

assay repeats.

0.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0

1 2 5

2 4 h o u rs


% c

ell

su

rv

iva

l

H N 5 p a re n ta l

E V -c lo n e 1

F la g -Y A P 1 -c lo n e 2*******

****

*

*

*

A


Flag-YAP1-clone 2 391.9 ± 58.8

EV-clone 1 212.6 ± 21.3


0.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0

1 2 5

4 8 h o u rs


% c

ell

su

rv

iva

l

H N 5 p a re n ta l

E V -c lo n e 1

F la g -Y A P 1 -c lo n e 2

** *****

**

*

*

B



EV-clone 1 22.6 ± 2.3


0.0

3.9

7.8

15.6

31.3

62.5

125.0

250.0

500.0

1000.0

0

2 5

5 0

7 5

1 0 0

1 2 5

7 2 h o u rs


% c

ell

su

rv

iva

l

H N 5 p a re n ta l

E V -c lo n e 1

F la g -Y A P 1 -c lo n e 2

* **

**

**

****

*

C



EV-clone 1 27.6 ± 2.8


130

4.3.5. Transient-over-expression of YAP1 in the non-cancerous COS-1 monkey

fibroblast cell line

This chapter so far demonstrates that over-expression of YAP1 promoted resistance to

reovirus oncolysis in 2 SCCHN cell lines. We consequently planned to over-express

YAP1 in COS-1, which is an African green monkey kidney fibroblast-like, simian

virus 40 (SV40) transformed, non-cancerous cell line. COS-1 cells are recognised as

easy transfection hosts. Therefore, we intended to check the transfection efficiency of

YAP1 in COS-1 cells, and to determine whether reovirus resistance associated with

over-expression of YAP1, is SCCHN cell line specific.

To over-express YAP1 in the COS-1 cell line, a transient transfection with the Flag-

tagged-YAP1 plasmid was performed (Section 2.16.1). Cells were also transfected

with the EV-control plasmid, or treated with 045.µL/well lipofectamine alone or with

media alone, which served as controls. At 24 hours post-transfection, COS-1 cell

lysates were collected and analysed for YAP1 protein expression by western blotting

(Section 2.14).

Figure 4.12 A demonstrates YAP1 protein over-expression in the cells transfected

with the Flag-YAP1 plasmid compared to cells treated with the EV-control plasmid,

lipofectamine alone or media alone. An exogenous YAP1 band at 56kDa was

detected in the cells treated with the Flag-YAP1 plasmid, and endogenous YAP1 was

detected in the samples at 54kDa. The combined intensity of the endogenous and

exogenous YAP1 bands in each sample was quantified and normalised to their

corresponding β-actin bands by densitometry analysis (Section 2.14.4) (Figure 4.12

B). A 21-fold increase in total YAP1 was observed in cells transfected with the Flag-

YAP1 plasmid compared to EV-control plasmid-transfected cells.

131

A

B

Figure 4.12. YAP1 protein expression in the COS-1 cell line after transient over-expression of

YAP1. Whole cell lysates were collected from cells treated with the Flag-YAP1 plasmid, media alone,

lipofectamine alone and the empty vector (EV)-control plasmid. A. YAP1 protein expression was

determined by western blotting. Endogenous YAP1 was detected in the samples at 54kDa, whereas

exogenous YAP1 was detected at 56kDa. Over-expression of YAP1 in the cells transfected with the

YAP1-plasmid was observed compared to the cells treated with the EV-plasmid, lipofectamine

transfection agent alone, or media alone. B. Total YAP1 protein expression in each sample was

normalised to the -actin loading control and quantified using densitometry. This confirmed that

transfection with the Flag-YAP1 plasmid caused a 21-fold increase in YAP1 expression with respect to

the EV-plasmid.

Fla

g-Y

AP

1

Med

ia o

nly

Lip

ofe

cta

min

e o

nly E

V

0

5

1 0

1 5

2 0

2 5

3 0

3 5


Re

lati

ve

de

ns

ity

132

4.3.6. Transient YAP1 over-expression caused increased resistance of the non-

cancerous COS-1 cell line to reovirus-induced cell death

Having confirmed transient YAP1 over-expression in the COS-1 cell line, the

susceptibility of these cells to reovirus oncolysis was investigated, in comparison to

cells treated with the EV-control plasmid, lipofectamine alone or media alone

(Section 2.16.3). The average OD of each un-infected transfection condition was

initially compared to check for cytotoxic effects. At 24 hours post-transfection, cells

were treated with media alone for an additional 48 hours before analysis via the MTS

assay (Section 2.9). There appeared to be some cytotoxicity in the cells transfected

with the Flag-YAP1 plasmid, compared to cells treated with media or lipofectamine

alone. However, minimal toxicity was observed in cells transfected with the EV-

control plasmid. This suggested that presence of YAP1 cDNA was responsible for the

cytotoxicity and was not simply due to cellular uptake of the vector (Figure 4.13).

Before infecting the cells with reovirus, the differences in transfection-associated

cytotoxicity was taken into account by counting cells from each treatment condition.

The cell counts were used to calculate the required MOI per well (Section 2.8), which

ensured that the same number of virus particles per cell was used in each treatment

condition. Cells were infected with serial dilutions of reovirus starting at MOI 250.

At 48 hours post-infection with reovirus, the % cell survival in each treatment

condition was assessed by the MTS assay (Section 2.9) and the IC50 values were

determined using CalcuSyn software (Section 2.27.4). Plasmid-mediated over-

expression of YAP1 in the COS-1 cell line caused a significant increase in resistance

to reovirus oncolysis at all MOIs tested (p<0.05 by un-paired t-test) (Figure 4.14).

Cells treated with lipofectamine alone, media alone or the EV-control plasmid had

IC50 values of MOI 6.7, MOI 11.1, and MOI 9.9 respectively. Compared to the EV-

control, there was a 10-fold increase in resistance to reovirus in cells transfected with

the Flag-YAP1 plasmid (IC50 MOI 100.8). This implied that over-expression of YAP1

may be a factor that promotes reovirus resistance in other cell types and is not a

phenomenon restricted only to SCCHN cell lines. Unlike the SCCHN cell lines, no

non-specific effect was caused by the EV-control plasmid in COS-1 cells. Although

this was an interesting result, in order to address our original study objectives, the

remaining parts of the project focussed on the effect of YAP1 expression in SCCHN.

133

Figure 4.13. The transfection-associated toxicity in COS-1 cells. After 24 hours transient-

transfection, COS-1 cells were treated with media alone for a further 48 hours before analysis via the

MTS assay. The raw Optical Density (OD) values for each treatment was compared. There was some

toxicity observed in the cells transfected with the Flag-YAP1 plasmid, but limited toxicity in cells

transfected with the empty vector (EV)-control plasmid. The graph shows the mean OD of 3 assay

repeats and error bars represent SD.

Med

ia o

nly

Lip

ofe

cta

min

e o

nly E

V

Fla

g-Y

AP

1

0 .0

0 .5

1 .0

1 .5


Op

tic

al

De

ns

ity

(4

90

nm

)

134

Figure 4.14. Plasmid-mediated over-expression of YAP1 increased the resistance of the COS-1

cell line to reovirus oncolysis. COS-1 cells were transiently transfected with the Flag-YAP1 plasmid

(red circles) or the empty vector (EV)-control plasmid (purple squares). Cells were also treated with

lipofectamine (green triangles) or media alone (blue triangles). After counting the cells in each

treatment condition, cells were subsequently infected with serial dilutions of reovirus, starting at MOI

250. The % cell survival in each treatment condition was then assessed using the MTS assay. The IC50

values of each treatment condition were determined using CalcuSyn software and are shown in a table

below each graph. Over-expression of YAP1 caused a 10-fold increase in the resistance to reovirus

oncolysis at all MOIs tested compared to the EV-control. *p<0.05, **p<0.01 and ***p<0.001 by un-

paired t-test. Error bars represent the SD from 2 assay repeats.

0.0

7.8

15.6

31.3

62.5

125.0

250.0

0

5 0

1 0 0


% c

ell

su

rv

iva

l

F la g -Y A P 1

E V


M e d ia o n ly

** *

***

** **

**




EV 9.9 ± 0.5

Flag-YAP1 100.8 ± 12.1

135

4.3.7. Cellular localisation of YAP1 in PJ34 and PJ41 SCCHN cell lines

The YAP1 protein has been shown to have dual functions in controlling cell

proliferation and apoptosis, depending on its cellular localisation [256]. Therefore,

the cellular location of YAP1 may be an important factor in understanding its

biological role in SCCHN cell lines and how it may impede reovirus oncolysis. There

are several compounds that are known to affect the function of YAP in the cell. In

order to select the most appropriate compound that alters the function of YAP1 and to

examine the resultant effect on reovirus oncolysis, it was necessary to perform an

immunofluorescent stain to establish the localisation of YAP1 in SCCHN cell lines.

The PJ41 and PJ34 SCCHN cell lines were permeabilised and then stained with a total

YAP1 primary antibody or a phospho-YAP primary antibody, which detected

endogenous levels of YAP only when phosphorylated at S127. A secondary antibody

conjugated to a fluorescent dye was then used to visualise the proteins by confocal

microscopy under the 488nm wavelength of light. A cell membrane marker, wheat

germ agglutinin (WGA), and a nuclear marker, TO-PRO-3, were also used to help

localise total YAP1 or phospho-YAP-S127 in the cells (Section 2.17).

Total YAP1 was predominantly expressed in the cytoplasm of the PJ41 cell line,

although there was also some expression in the nucleus, as demonstrated by the

intensity profile (Figure 4.15 and Figure 4.16 A). As expected, phospho-YAP-127

was exclusively expressed in the cytoplasm of PJ41 cells (Figure 4.15 and Figure

4.16 B). This implied that the majority of YAP1 was being phosphorylated by

upstream components of the Hippo pathway to sequester it to the cytoplasm where it

remains in-active, but some YAP1 protein was not phosphorylated and was possibly

having an oncogenic function in the nucleus. Total YAP1 and phospho-YAP-S127

were un-detectable in the PJ34 cell line. The staining intensity of the YAP1 protein

in these cell lines was consistent with the mRNA expression data obtained in Chapter

3, Section 3.3.3, where PJ41 displayed highest YAP1 and PJ34 showed lowest YAP1

expression. Total YAP1 and phospho-YAP-S127 staining was performed on different

days, but the two cell lines were stained simultaneously and imaged using the same

confocal settings. Thus, direct comparisons could be made between the cell lines, but

there was variation in the staining intensity between total YAP1 and phospho-YAP-

136

S127. Despite this, the cellular localisation of YAP1 and phospho-YAP-S127 could

be compared.

Figure 4.15. Immunofluorescent staining of PJ41 and PJ34 SCCHN cell lines for the YAP1

protein. All images were taken using the confocal microscope at ×40 magnification. Total YAP1 and

phospho-YAP-S127 (pYAP-S127) were detected predominantly in the cytoplasm of permeabilised

PJ41 cells, as shown by green staining, but were un-detectable in permeabilised PJ34 cells. There was

no YAP1 or pYAP-S127 staining found in permeabilised cells treated only with secondary antibody,

which served as a negative control for each cell line (inset, top right). Wheat germ agglutinin (WGA)

was used as a cell membrane marker (red stain) and TO-PRO-3 was used to detect the cell nucleus

(blue stain).

137

A

B

Figure 4.16. Intensity profiles of total YAP1 and phospho-YAP-S127 in the PJ41 cell line. The

intensity profiles were measured in PJ41 cells stained with A. total YAP1 and B. pYAP-S127. The

white arrow shows where the 3 different fluorescent labels were expressed through the cells. The blue

peak represented the nucleus, the red peak indicated the cell membrane, and the green intensity peak

demonstrated the presence of total YAP1 or pYAP-S127. Total YAP1 was expressed predominantly in

the cytoplasm, but there was also some YAP1 detected in the nucleus. pYAP-S127 was expressed

solely in the cytoplasm. Images were taken using the confocal microscope at ×40 magnification.

138

4.3.8. Treatment of the PJ41 cell line with Sphingosine-1-phosphate (S1P)

caused sensitisation to reovirus oncolysis

As the YAP1 protein was localised predominantly in the cytoplasm of the PJ41 cell

line in its phosphorylated state, we intended to force YAP1 into the nucleus using

sphingosine-1-phosphate (S1P) and then assess the effect on reovirus-induced cell

death (Section 2.18). S1P has been shown to cause de-phosphorylation of YAP on

residue serine 127 (S127), resulting in nuclear migration of YAP in human embryonic

kidney or breast epithelial cell lines [220].

First, the toxicity associated with S1P treatment was taken into account by counting

cells with or without S1P treatment. The cell counts were used to calculate the

required MOI per well (Section 2.8), which ensured that the same number of virus

particles per cell was used in each treatment condition. PJ41 cells were treated with

1µM S1P for 60 minutes and then infected with serial dilutions of reovirus, starting at

MOI 500, for 24 hours. Cells were also treated with S1P alone (without reovirus

infection) or with reovirus alone (without S1P treatment). The % cell survival was

determined by the MTS assay (Section 2.9) and the IC50 values were evaluated using

CalcuSyn software (Section 2.27.4). S1P treatment caused minimal toxicity

compared to PJ41 cells treated with media alone (Figure 4.17 A). S1P treatment

sensitised PJ41 cells to reovirus-induced cell death by 10-fold (IC50 = MOI 1346.59)

compared to cells treated with reovirus alone (IC50 = MOI 132.26) (Figure 4.17 B).

The ability of S1P to induce de-phosphorylation of YAP1 in the PJ41 cell line was

then assessed. According to Yu et al, de-phosphorylation of YAP was detected in the

human embryonic kidney HEK293A cell line treated with S1P at a concentration of

1µM for 60 minutes. Treatment with S1P at time-points longer than 60 minutes

showed that the de-phosphorylation was less effective [220]. Therefore, we treated

PJ41 cells with media alone, or with 1µM S1P for 20, 30 and 60 minutes. We also

treated HEK293A cells in the same way for use as a positive control (Section 2.18).

Cell lysates were then collected and phospho-YAP-S127 or total YAP1 levels were

detected by western blotting (Section 2.14). Western blot analysis showed little

quantifiable reduction in phosphorylated YAP after S1P treatment in the PJ41 cell

line, compared to the un-treated control cells (Figure 4.18 A and B). As expected,

total YAP1 levels remained relatively constant after S1P treatment. Unlike results

139

published by Yu et al, we did not detect any total YAP1 or phospho-YAP-S127 in

HEK293A cell lysates treated with or without S1P. We did increase the protein

concentration of HEK293A lysates loaded on to the SDS-PAGE gel in the hope of

detecting positive bands, but this was also unsuccessful. Altogether, these results

implied that S1P promoted reovirus oncolysis via a different mechanism to the de-

phosphorylation of YAP1, as discussed later in this chapter.

140

A

B

Figure 4.17. Treatment of the PJ41 SCCHN cell line with S1P caused sensitisation to reovirus-

mediated cell death. A. PJ41 cells were treated with 1µM S1P for 60 minutes and then replaced with

fresh growth media for a further 24 hours. Cells were also treated with media alone for the same

duration of time before analysis via the MTS assay. The raw Optical Density (OD) values showed that

S1P caused little cytotoxicity. The graph shows the mean OD of 2 assay repeats and error bars

represent SD. B. PJ41 cells were treated with 1µM S1P (red circles) or with media alone (blue

triangles) for 60 minutes, counted, and were then infected with serial dilutions of reovirus for 24 hours,

starting at MOI 500. Cells were also treated with 1µM S1P for 60 minutes, and then treated with media

alone (without reovirus infection) for a further 24 hours (dotted grey line). The % cell survival in each

treatment group was determined by the MTS assay. S1P treatment sensitised PJ41 cells to reovirus-

induced cell death by 10-fold compared to cells treated with reovirus and media alone. The IC50 values

were determined using CalcuSyn software and are shown in a table below the graph. *p<0.05 and

**p<0.01 by un-paired t-test. Error bars represent the SD from 2 assay repeats.

0.0

15.6

31.3

62.5

125.0

250.0

500.0

0

2 5

5 0

7 5

1 0 0


% c

ell

su

rv

iva

l

S 1 P (1 µ M ) + R e o v iru s

S 1 P (1 µ M ) + m e d ia a lo n e

m e d ia a lo n e + R e o v iru s

***

*

** ***


Media alone + Reovirus 1346.59 ± 80.8

S1P (1µM) + Reovirus 132.26 ± 7.9

S1P

(1µM

)

Med

ia o

nly

0 .0

0 .1

0 .2

0 .3

0 .4

0 .5

Op

tic

al

De

ns

ity

(4

90

nm

)

141

A

B C

Figure 4.18. Treatment of the PJ41 SCCHN cell line with S1P failed to de-phosphorylate YAP.

Whole cell lysates were collected from un-treated cells, or cells treated with 1µM S1P for 20, 30, or 60

minutes. A. pYAP-S127 and total YAP1 protein expression was determined by western blotting.

pYAP-S127 was detected in the samples at 65kDa, whereas total YAP1 was detected at 54kDa. B.

pYAP-S127 and C. total YAP1 protein expression in each sample was normalised to the -actin

loading control and quantified using densitometry. There was little de-phosphorylation of YAP

observed in the cells treated with S1P compared to un-treated cells. Total YAP1 expression remained

almost unchanged after S1P treatment.

un

treate

d

S1P

(1µM

) 20 m

ins

S1P

(1µM

) 30 m

ins

S1P

(1µM

) 60 m

ins

0

2 0

4 0

6 0

8 0

1 0 0

Y A P 1

Re

lati

ve

de

ns

ity

un

treate

d

S1P

(1µM

) 20 m

ins

S1P

(1µM

) 30 m

ins

S1P

(1µM

) 60 m

ins

0

5 0

1 0 0

1 5 0

p Y A P -S 1 2 7

Re

lati

ve

de

ns

ity

142

4.4. DISCUSSION

Knock-down of YAP1 by siRNA-mediated transfection caused significant

sensitisation of the PJ41 SCCHN cell line to reovirus oncolysis (Chapter 3). The aim

of this chapter was to explore the effect of YAP1 over-expression on reovirus

oncolysis in SCCHN cell lines. To test this, YAP1 was transiently over-expressed in

the PJ34 SCCHN cell line, as it was the most sensitive to reovirus oncolysis and

displayed the lowest expression of the YAP1 gene. Stable-YAP1-over-expressing cells

were also generated from the HN5 SCCHN cell line, before infection with reovirus

and assessment of the effect on cell survival. Transient over-expression of YAP1 in

the non-cancerous COS-1 monkey fibroblast cell line was also performed in order to

evaluate whether YAP1 expression affects the susceptibility of different cell types to

reovirus oncolysis. Furthermore, the plan was to use a compound that stimulated the

de-phosphorylation and nuclear localisation of the YAP1 protein in the PJ41 SCCHN

cell line, and subsequently test the efficiency of reovirus oncolysis.

Our results showed that plasmid-mediated over-expression of YAP1 in the PJ34 cell

line caused enhanced resistance to reovirus oncolysis at all MOIs tested compared to

the negative controls. The reovirus IC50 of the cells transfected with the EYFP-YAP1

and Flag-YAP1 plasmids were MOI 77.6 and MOI 27.9 respectively. This

corresponded to a 12- and 4-fold increase in resistance to reovirus treatment compared

to cells treated with lipofectamine or media alone (IC50 values = MOI 6.7). However,

it is important to note that empty-vector (EV) controls (i.e. the plasmids without the

YAP1 insert) were not included in these experiments. It is therefore probable that

over-expression of YAP1 induces reovirus resistance in the PJ34 cell line, but we

cannot confidently conclude this without knowing the effect of non-independent

variables caused by an EV. We then used stable transfection as a tool to study the

effect of increased YAP1 expression on the cellular physiological response to reovirus

infection over a time period of 72 hours. Attempts to create stable-YAP1-over-

expressing clones from the PJ34 SCCHN cell line were unsuccessful. The HN5

SCCHN cell line proved to be a more efficient transfection host than PJ34 and

therefore, stable-YAP1-over-expressing clones were created from parental HN5 cells.

Out of the 3 SCCHN cell lines tested in this study, HN5 cells were the next cell line of

choice because they were more sensitive to reovirus-induced cell death and had lower

143

endogenous expression of YAP1 than the PJ41 SCCHN cell line. Initially, the reovirus

IC50 value of stable-EYFP-YAP1-clone 6 was evaluated against the IC50 of the HN5

parental cell line. This revealed that EYFP-YAP1-clone 6 (which showed a 5-fold

over-expression of the YAP1 protein) was 8-, 6- and 3-fold more resistant to reovirus

than the HN5 parental cell line at 24, 48 and 72 hours post infection respectively.

This was an interesting result, but caution must be taken since an EV-control plasmid

was not included for direct comparison. To address this issue, later on in the project,

we acquired a pcDNA3.1 EV control plasmid, which contained the same components

as the Flag-YAP1 plasmid, but without the YAP1-tag insert. Stable clones were made

from HN5 parental cells transfected with the EV and Flag-YAP1 plasmids. Out of all

the clones tested, Flag-YAP1-clone 2 exhibited a 4-fold increase in total YAP1 protein

expression. As expected, the EV-clones did not over-express YAP1 compared to

HN5 wild-type cells. Subsequent infection with reovirus revealed that there was some

non-specific-resistance caused by the EV-clone compared to HN5 parental cells, but

this effect became less apparent at later times of infection. Compared to EV-clone 1

stable cells, Flag-YAP1 clone 6 cells were 2-fold more resistant to reovirus at 24

hours, and 3-fold more resistant at 48 and 72 hours post-infection. This was

statistically significant at MOIs between 7.8 and 62.5 at all time-points tested, and the

IC50 values of the HN5 parental cell line lied within this range. There were no

statistical differences in % cell survival between the EV and Flag-YAP1 stable cell

lines treated with reovirus at higher MOIs, probably because a high proportion of cells

were non-viable. Overall, our results implied that the expression level of YAP1 is

important in determining the degree of reovirus resistance in the HN5 SCCHN cell

line, and this paralleled the result of the transient over-expression of YAP1 in the PJ34

SCCHN cell line.

It was interesting to discover that stable over-expression of the YAP1 protein in HN5

cells caused an increase in cell growth rate compared to stable EV-control cells or the

HN5 parental cell line. Over-expression of YAP1 also stimulated cell proliferation in

non-small-cell lung cancer (NSCLC), hepatocellular carcinoma and endometrial

cancer cell lines [219, 293, 294]. YAP1 is a key regulator of organ size by

orchestrating cell proliferation and apoptosis, and is a key down-stream target of the

Hippo signalling pathway [219, 273, 295]. The cellular localisation of YAP has been

shown to predict its function. For example, core serine/threonine kinases of the Hippo

144

signalling pathway phosphorylate downstream YAP on serine 127 (S127)) [251],

which leads to the cytoplasmic retention and inactivation of YAP. Cell proliferation

is therefore prevented [252]. In the absence of Hippo signalling, YAP migrates to the

nucleus where it acts as a cofactor to stimulate expression of genes that promote

proliferation and inhibit apoptosis [253, 275]. Further work would determine the

cellular localisation of YAP1 following plasmid-mediated over-expression, and after

reovirus infection. Immunofluorescent staining for total YAP1 and cellular

fractionation followed by immunoblotting would provide evidence for this. One

would assume that because over-expression of YAP1 caused an increase in cell

growth rate, that some exogenous YAP1 would be observed in the nucleus. However,

YAP1 can be expressed in the cytoplasm and the nucleus at the same time [252], and

molecular pathways may not be fixed but may actively change depending on the

context and upstream input [219, 296]. An explanation for our findings may be that

over-expression of YAP1 provides a survival advantage to HN5 cells by functioning

simultaneously as part of an anti-viral response pathway and as a growth promoter,

and can shuttle between the cytoplasm and the nucleus to meet the needs of the cell.

Since the cellular location of YAP1 may be an important factor in understanding how

it may impede reovirus oncolysis in SCCHN, an immunofluorescent stain was

performed in the PJ41 cell line. The PJ41 cell line was selected because it displayed

the highest YAP1 expression and was the most resistant SCCHN cell line to reovirus-

induced cell death (Chapter 3). Endogenous YAP1 was predominantly localised in

the cytoplasm of PJ41 cells in its phosphorylated state. Thus, we attempted to force

YAP1 into the nucleus using sphingosine-1-phosphate (S1P) and then assess whether

this affected the ability of reovirus to induce cell death. S1P has been shown to signal

through the Gα proteins G12/13 to activate Rho and the actin cytoskeleton, which

inhibits the core kinases, LATS 1 and 2 of the Hippo signalling pathway. This in turn

has been shown to cause de-phosphorylation of YAP at S127, resulting in nuclear

migration of YAP, enhanced target gene expression and cell proliferation in various

different cell lines [220]. We were unable to detect significant de-phosphorylation of

YAP by S1P treatment in the PJ41 cell line, despite repeating the experiment using

fresh lysates. However, we unexpectedly observed a 10-fold sensitisation to reovirus

oncolysis in PJ41 cells treated with S1P (IC50 = MOI 1346.59) compared to PJ41 cells

treated with reovirus alone (IC50 = MOI 132.26). This suggested that S1P promotes

145

reovirus oncolysis via a different mechanism than the de-phosphorylation and nuclear

migration of YAP. It is therefore still uncertain whether the cellular localisation of

YAP influences the fate of the cell after reovirus infection. The fact that we did not

detect total YAP or pYAP-S127 in HEK293A cell lysates made it difficult to gauge

the repeatability of the experiment and perhaps the PJ41 cell line, being of different

origin, needed longer treatment with S1P in order to notice adequate de-

phosphorylation of YAP. Therefore, it is possible that the de-phosphorylation

occurred later than 60 minutes and after the addition of reovirus to the cells, which

may justify the observed S1P-induced sensitisation to reovirus oncolysis.

Lysophosphatidic acid (LPA) and thrombin have also been shown to promote the

nuclear accumulation of YAP through interaction with upstream Hippo signalling

proteins [220, 297]. It may be of value to ascertain whether these small molecules can

alter the cellular localisation of YAP1 in the PJ41 SCCHN cell line and subsequently

evaluate the resultant effect on reovirus oncolysis. It may also be meaningful to test

these molecules on the stable-YAP1-over-expressing HN5 cells to see whether the

level of resistance to reovirus changes in comparison to the EV-control stable cell

line. There are other known modulators of YAP that stimulate its non-oncogenic

function, i.e. its phosphorylation, inactivation and cytoplasmic retention. These

include verteporforin [298], dobutamine [273], latrunculin A [299] and dasatinib

[300]. It seemed more logical to use a compound that promoted the nuclear

localisation of YAP1, as we found it to be mainly expressed in the cytoplasm and was

phosphorylated at residue S127 in the PJ41 cell line. However, as some nuclear un-

phosphorylated YAP1 was also detected, it may be interesting to investigate what

effect the latter molecules have on reovirus oncolysis.

Although we have identified the YAP1 protein to influence reovirus oncolysis in

SCCHN cell lines, there are clearly other unknown factors that contribute to this.

This is not surprising, considering the interconnectivity between molecular signalling

pathways that tightly control cellular growth, proliferation, differentiation and death,

and not to mention the biological heterogeneity of SCCHN. It is important to note

that S1P specifically targets the upstream kinases of the Hippo pathway. Although

this pathway seems to be the dominant regulator of YAP, components from other

molecular pathways also have the ability to control its function. Akt (also known as

protein kinase B (PKB)) is a serine/threonine kinase that can phosphorylate YAP at

146

S127 independently of the Hippo pathway, leading to its cytoplasmic retention [301].

Cross-talk between the WNT/-catenin [302], TGF [303], GPCR [220] signalling

and Sonic hedgehog (Shh) [304] signal transduction pathways can cooperate with the

Hippo pathway to control cell growth and proliferation [278]. Therefore, it is possible

that the phosphorylated-S127 portion of YAP that we visualised in the cytoplasm of

PJ41 cells is not due to upstream Hippo signalling. If this was the case then this may

explain why we did not detect de-phosphorylation of YAP1 by S1P treatment. It

would be interesting to confirm this by monitoring Hippo signalling activities in the

SCCHN cell lines by using a PCR array, which allows the profile of nearly 400

related genes to be measured at the same time. Even if Hippo signalling is not the key

regulator of YAP1 in these cells, if other up-stream YAP1-related genes could be

identified, then several genes could be knocked-down or over-expressed all at once. It

is uncertain whether altering the expression of multiple genes at the same time would

be detrimental to cell, but it would be interesting to determine whether this would

enhance the effect on reovirus oncolysis, as oppose to targeting YAP1 on its own.

Another factor to consider is the mammalian YAP co-protein, TAZ (see Figure 4.1).

TAZ has a similar molecular structure and function to YAP and contains

serine/threonine residues that can be phosphorylated by upstream Hippo kinases.

Like YAP, TAZ binds to transcription factors in the nucleus to stimulate expression of

growth promoting genes. YAP is a relatively stable protein that is mainly regulated

by cytoplasmic-nuclear shuttling. Unlike YAP however, TAZ is a very unstable

protein that has a short half-life of less than two-hours, which suggests that the main

path of TAZ inhibition is through protein degradation [276]. TAZ was not one of the

genes selected in our original DNA microarray screen, but it would be intriguing to

see whether the simultaneous knock-down of YAP and TAZ augments the cells’

susceptibility to reovirus oncolysis even more so than knock-down of YAP1 alone.

S1P is a downstream product of sphingolipids, which are bioactive lipid mediators

[305]. S1P is an extracellular ligand for sphingosine-1-phosphate receptor 1 (S1PR1),

a G-protein coupled receptor, and is a key regulator of the immune and vascular

systems. S1P-cell surface receptor signalling regulates angiogenesis, permeability,

vascular stability and the trafficking of T- and B- cells from lymphoid organs into the

lymphatic vessels [306]. S1P regulation has been shown to drive tumorigenesis and

neovascularisation [307]. The effects of S1P on host cell defences against virus

147

infection is not well understood, but over-expression of sphingosine kinase-1 (SK1)

(which converts sphingosine to S1P) increased the susceptibility of human HEK293

embryonic kidney cells to influenza virus infection [305]. In a similar way, perhaps

S1P treatment in PJ41 SCCHN cells heightened the cells’ susceptibility to reovirus

infection.

COS-1 is an SV40-transformed monkey fibroblast-cell line, and is an amenable

transfection host. Fibroblasts derived from other species are often used to study the

function of human genes and their protein products [132, 133]. Our findings

demonstrated a 21-fold over-expression of the YAP1 protein in COS-1 cells after

plasmid-mediated transfection. This consequently caused a 10-fold increase in

resistance to reovirus oncolysis compared to EV-control treated cells. This implied

that over-expression of YAP1 may be a universal factor that promotes reovirus

resistance in other cell types and is not only associated with SCCHN cell lines. This

would be worth further investigation and is something to consider for future work.

However, in order to address our original study objectives, we concentrated on the

effect of YAP1 expression in SCCHN. Transfection with the EV-control plasmid had

little effect on cell survival after reovirus infection in the non-cancerous COS-1 cell

line, but caused a certain amount of non-specific resistance to reovirus in the SCCHN

HN5 cell line. This may have been a consequence of chromosomal instability caused

by the combined dysfunctional effects of oncogenes and tumour suppressor genes in

tumour cells [308], which would not be present in a non-cancerous genome.

Therefore, integration of the EV-control plasmid in HN5 cells may have caused new

mutations that affected anti-viral-related host genes, giving rise to a more resistant

phenotype. It is also worthy to note that the reovirus IC50 value of COS-1 cells at 48

hours post-infection was MOI 11.1. In contrast to the mean IC50 values calculated at

the same time-point (Chapter 3, Section 3.3.2), transformed-COS-1 cells were more

susceptible to reovirus oncolysis than the un-transformed MRC-5 human lung

fibroblast cell line (IC50 = MOI 2769.7), and PBMCs isolated from the blood of a

human healthy donor (IC50 = MOI 1509.6). This concurs with what is documented in

the literature, that transformed cell lines are more susceptible to reovirus-induced cell

death than un-transformed cell lines [124, 125]. In relation to the SCCHN cell lines,

COS-1 cells had approximately equal reovirus sensitivity to HN5 cells at 48 hours

post-infection.

148

4.5. CONCLUSION

Stable over-expression of YAP1 in the HN5 SCCHN cell line significantly increased

the resistance to reovirus oncolysis. Transient over-expression of the YAP1 protein

likely contributes to enhanced reovirus resistance in the PJ34 SCCHN cell line, and

undoubtedly does so in COS-1 monkey kidney fibroblast cells. Our attempts to

detect de-phosphorylation of YAP1 by S1P in the PJ41 SCCHN cell line was

unsuccessful, but treatment with S1P did consequently sensitise cells to reovirus

oncolysis. It is unclear how S1P mediates this effect as it appears to act

independently of YAP, which emphasises the complexity of the cellular response to

reoviral infection in the PJ41 cell line. Overall, our results implied that the level of

YAP1 expression is important in determining the susceptibility of SCCHN cell lines,

and possibly other cell lines of different origin, to reovirus-induced cell death.

Understanding how YAP1 facilitates this in SCCHN could potentially lead to its use

as a biomarker of treatment response in clinical trial patients receiving reovirus

therapy, and will therefore be studied in more detail in Chapter 5. The expression of

YAP1 will also be analysed in human head and neck carcinoma tissues compared to

normal tissue samples.

149

CHAPTER 5

MECHANISTIC STUDIES BEHIND THE

INFLUENCE OF YAP1 ON REOVIRUS

ONCOLYSIS IN SCCHN CELL LINES

150

5. MECHANISTIC STUDIES BEHIND THE INFLUENCE OF YAP1 ON

REOVIRUS ONCOLYSIS IN SCCHN CELL LINES

5.1 INTRODUCTION

Reovirus T3D displays great promise as an anti-cancer therapeutic. The mechanism

of reovirus oncolysis has been a controversial subject and still remains to be fully

elucidated. Understanding this process is important as it could lead to the discovery

of new biomarkers of reovirus treatment response and thus, improve clinical trial

design and patient selection. The most recognized model is based on the idea that

reovirus exploits aberrant Ras signalling pathways in cancer cells. However, it is

becoming widely accepted that this is not always the case. Ras-transformed tumour

cells can actually develop resistance to reovirus-induced cytotoxicity [309]. In

particular, no link between activated Ras signalling and reovirus oncolysis in SCCHN

cell lines was found [145]. This finding formed the rationale for this project; to test a

panel of target genes that may be involved in, or predict for, reovirus-mediated

oncolysis. We have so far established that host-cell expression of the YAP1 protein

effects the susceptibility of SCCHN cell lines to reovirus-induced cell death. Finding

the mechanism of how YAP1 mediates this effect is the purpose of this chapter, and

hence, it is imperative to discuss the known factors that can affect reovirus oncolysis.

In addition to the components of Ras signalling, abnormalities in other signalling

pathways may affect the oncolytic tropism of reovirus. Both oncogene activations

and inactivation of tumour suppressor genes contribute to carcinogenesis.

Dysfunction of p53, ataxia telangiectasia mutated (ATM) and pRb tumour suppressor

genes can increase genomic instability and disturb cell cycle control, apoptotic

signalling and intact interferon responses to viral infection [310, 311]. It has been

demonstrated that inactivating mutations in these tumour suppressor genes enhanced

the susceptibility of human cancer cells to oncolytic viruses, including adenovirus,

myxoma virus and reovirus, compared to cancer cells with normal p53, ATM and pRb

activity [310, 312]. Another factor known to affect reovirus oncolysis is cell cycle

phase. Heinemann et al observed an increased sensitivity of B16.F10 mouse

melanoma cells to reovirus-induced cell death after treatment with hydroxyurea, a cell

synchronizer, which correlated with increased viral replication [313].

151

Apart from direct viral replication, reovirus can induce cancer cell death in different

ways and thus, there are many potential pathways that YAP1 could be affecting.

Firstly, programmed necrotic cell death (necroptosis) can be initiated by reovirus

infection, and is induced by binding of agonists such as tumour necrosis factor-related

apoptosis-inducing ligand (TRAIL), tumour necrosis factor-α (TNF-α) and Fas ligand

(FasL), to death receptors on the cell surface, including TNFR1, TNFR2 and Fas [314,

315]. This results in downstream regulation of receptor interacting protein 1 (RIP1)

or 3 (RIP3), which are hallmark mediators of necroptosis [314-316]. Other known

mediators of necroptosis include cylindromatosis (CYLD), TNF receptor-associated

factors (TRAFs), adenine nucleotide translocase (ANT), poly ADP-ribose

polymerases (PARPs) and reactive oxygen species (ROS) [315, 316]. Although the

signalling pathways leading to necrosis and necroptosis are evidently separate, there

are no distinct morphological differences [317]. Like all forms of necrotic cell death,

necroptosis is characterised by an increase in cell volume, swelling of organelles,

rupture of the cell membrane and leakage of intracellular contents [317]. Significant

necrosis was identified in xenograft human SCCHN specimens treated with

intratumoral injections of reovirus, with no signs of apoptosis [235]. Recent evidence

suggests that there is substantial interplay between necroptosis and apoptosis, as some

proteins are shared between both signalling pathways [316].

Secondly, autophagy has been demonstrated to be yet another mode of reovirus-

induced cell death. Autophagy is a regulated cellular process in eukaryotic cells that

functions to deliver cytoplasmic organelles, proteins and macromolecules to the

lysosome for degradation and recycling. Reovirus infection of human multiple

myeloma cell lines caused an induction of autophagy, which was suppressed by the 3-

methyladenine (3-MA) autophagy inhibitor [318]. Endoplasmic reticular (ER) stress

triggered by Akt-mTOR signalling has been shown to induce autophagy [315, 319].

Since recent studies have demonstrated that reovirus infection can stimulate ER-

stress-induced apoptosis, it is reasonable to predict that ER signalling can also cause

autophagic cell death during mammalian reovirus oncolysis. Supporting this theory,

infection of Vero cells with avian reovirus induced autophagy through

PI3K/Akt/mTOR signalling [320].

152

Thirdly, apoptosis is a major mechanism of reovirus-mediated cell death. The

morphological hallmarks of apoptosis are cell membrane blebbing, cell shrinkage,

chromatin condensation and nuclear fragmentation. Both the S1 and M2 gene

segments of reovirus T3D play important roles in reovirus-induced apoptosis [243].

S1 encodes the σ1 protein that is important for virus cell attachment, whereas M2

encodes the µ1/µ1c outer capsid protein that is cleaved during proteolytic disassembly

of virions to form ISVPs, suggesting that these are key processes needed for the

initiation of reovirus-induced apoptosis [243, 321, 322]. There is a high level of

cross-talk between the intrinsic (mitochondrial-related) and extrinsic (death receptor-

related) apoptotic signalling pathways during reovirus-induced apoptosis [315, 323].

After infection, TRAIL ligands can bind to cell surface death receptors (DRs), DR4

and DR5, resulting in the recruitment of Fas-associated death domain (FADD) (an

adaptor molecule) and pro-caspase-8 and -10 to form the death-inducing signalling

complex (DISC). Pro-caspase-8 then becomes cleaved and activated, which in turn

activates caspase-3 to induce the final apoptotic execution pathway [243, 323-325].

Additionally, reovirus infection can stimulate mitochondrial signalling. Activation of

caspase-8 can induce the cleavage of Bid, a pro-apoptotic BH3-only Bcl-2 family

protein. Truncated Bid then migrates to the mitochondria and disrupts the interactions

between pro-apoptotic (Bax and Bak) and anti-apoptotic (Bcl-2 and Bcl-xL) proteins,

which forms a pore in the mitochondrial membrane, allowing the release of pro-

apoptotic proteins cytochrome c and second mitochondrion-derived activator of

caspases (Smac) [242, 243]. Release of these proteins activates caspase-9 that then

activates effector caspases-3 and -7 to induce apoptosis [242, 243]. Reovirus

infection has been shown to activate the transcription factor nuclear factor kappa B

(NF-κB) to induce apoptosis by stabilization of the p53 tumour suppressor protein

[229] or by upregulation of TRAIL and DR expression [243]. Noxa, another pro-

apoptotic Bcl-2 family member, has an important role in reovirus-induced apoptosis,

and its expression is dependent on NF-κB and interferon-regulatory 3 (IRF-3)

transcription factor activity [326], but can be activated independently of interferon-

beta (IFN-β). Activation of c-jun N-terminal kinase (JNK) and c-jun, a JNK-

associated transcription factor [327], have been connected to reovirus-induced

apoptosis, as has increased expression of ER-stress-related genes such as GADD34,

CHOP, GRP78 and the spliced form of XBP-1, in infected pancreatic cancer cell lines

[228]. The purpose of the endoplasmic reticulum is to fold and process secretory and

153

transmembrane proteins. A balance between ER protein load and the capacity to

process must be achieved to enable the correct folding of proteins [328]. Certain

stimuli such as viral infection can disrupt the normal function of the ER to cause a

build-up of misfolded and unfolded proteins. This is coined ER stress. In an attempt

to reduce ER stress, the unfolded protein response (UPR) becomes activated [328]. If

however, normal conditions are not restored, ER stress can cause cell death by

apoptosis, as demonstrated in reovirus infected human muliple myeloma [329] and

mutant melanoma cell lines [330].

Another critical factor that contributes to the efficiency of reovirus oncolysis in both

in vitro and in vivo systems is the immune system. Human Type III interferons (IFN),

namely IFN-1, IFN-2 and IFN-3, are secreted in response to viral infection but

their exact role remains to be determined. The Type II IFN- cytokine is not

produced in response to virus infection [331]. Human type I interferons are a family

of cytokines that are a major part of the innate response against virus infection, and

are comprised of one IFN-β and thirteen IFN-α members. IFN-ώ, IFN-ε, IFN-τ, IFN-

δ, and IFN-κ cytokines are also members of this family but have less important roles

in anti-viral responses [331]. The type I IFN response is initiated when a virus

produces dsRNA during replication. Cell sensors called pattern recognition receptors

(PRR), including toll-like receptors (TLR), retinoic acid inducible gene-1 (RIG-I),

melanoma differentiation-associated protein-5 (MDA5) and double-stranded RNA-

activated protein kinase (PKR), recognize and bind to the viral dsRNA, leading to

activation of transcription factors NF-κB and IRF-3 [331-333]. This results in the

expression and secretion of type I IFN- and IFN- that bind to the IFN-/ receptor

(IFNAR) to activate janus kinase-1 (JAK1) and tyrosine kinase-2 (Tyk2), which

phosphorylate and activate transcription factors, signal transducers and activators of

transcription (STAT)-1 and -2 [331, 332]. The STAT1-STAT2 complex associates

with a third transcription factor, interferon regulatory factor-9 (IRF-9), to form a

heterotrimeric transcription factor complex (ISGF3), which migrates to the nucleus to

bind to IFN-stimulated response elements (ISREs) that are present in the promoters of

most IFN-responsive genes. This interaction stimulates the transcription of anti-viral

IFN-stimulated genes (ISGs) [331, 332], resulting in inhibition of viral replication.

Shmulevitz et al showed that compared to non-transformed cells, Ras-transformed

NIH-3T3 fibroblasts inhibited the expression of certain ISGs and IFN- to enhance

154

reoviral spread and oncolysis [333]. Cancer cells are less able to respond to IFN than

normal cells, which may partly account for effective reovirus oncolysis [334].

Reovirus infection of tumour cells can also generate adaptive immune responses.

There is concern that systemic delivery of reovirus is hindered by B-cell production of

neutralising anti-reovirus antibodies (NARA) before it reaches the site of the tumour

[335]. However, anti-tumour activity was still notable after intravenous administration

of reovirus in patients with advanced cancers, despite finding substantial neutralizing

anti-reoviral antibody titers [169, 174]. Reovirus may evade NARA by attaching to

circulating blood cells that carry, transport and protect this virus [179, 225, 336].

Alternatively, animal models have demonstrated that infection of tumour cells with

oncolytic viruses may facilitate priming of an anti-tumour response. For example,

intratumoral injection of reovirus altered the immune milieu of the tumour

microenvironment in a melanoma xenograft in vivo model, and was associated with

the release of inflammatory cytokines and chemokines, including interleukin-6 and -8

(IL-6 and IL-8) to promote tumour cell killing [335].

The Hippo signalling pathway functions to control organ size by modulating apoptosis

and cell proliferation, but the upstream regulation of this pathway is not well

understood. To our knowledge, there is no published data directly linking

Hippo/YAP1 signalling to reovirus infection, although there are potential connections

that may be worth further exploration. Firstly, the regulation of apoptosis by YAP1

could be affecting the efficiency of reovirus oncolysis, as mentioned in the discussion

of this chapter. Secondly, the actin cytoskeleton is an upstream regulator of the Hippo

pathway [337], and reovirus can stabilise cellular microtubules to aid its replication

[123, 191, 192], suggesting that cytoskeletal components are possible influences.

Thirdly, the main reovirus receptor, JAM-A, interacts with PDZ-domain containing

proteins such as ZO-2 [267]. YAP also contains a PDZ-domain and the YAP2

isoform has been shown to complex with ZO-2 at tight junctions [266]. It could

therefore be hypothesised that YAP1 mediates resistance at the cell surface to prevent

reovirus cell entry. We aim to investigate how host-cell expression of YAP1 affects

reovirus oncolysis, whether it be through the possible control of viral entry, viral

replication, innate anti-viral immune responses, the cell cycle, necroptosis, autophagy

or apoptosis.

155

5.2 STUDY OBJECTIVE

The objective of this chapter was to investigate how YAP1 expression influences

reovirus-mediated oncolysis in SCCHN cell lines.


1. Measurement of YAP1 protein levels in SCCHN cell lines and in stable-YAP1-

over-expressing cell lines pre- and post-infection with reovirus by flow cytometry

analysis, to determine any changes in YAP1 expression after reovirus infection.

2. Measurement of JAM-A protein levels in SCCHN cell lines and in stable-YAP1-

over-expressing cell lines by flow cytometry analysis, to determine whether

reovirus entry is restricted at the cell surface.

3. Immunofluorescent staining and confocal imaging of reovirus protein after

infection of SCCHN cell lines and stable-YAP1-over-expressing cell lines, to

assess whether reovirus entry is inhibited at the cell surface membrane.

4. Measurement of intracellular and extracellular reovirus protein production in

infected SCCHN cell lines and in stable-YAP1-over-expressing cell lines, by

TCID50 assay, western blotting or flow cytometry analysis. This would determine

whether the rate of reovirus replication or release are being affected.

5. Measurement of IFN-β secretion in SCCHN cell lines and in stable-YAP1-over-

expressing cells infected with reovirus by the Verikine™ Human IFN-β ELISA

kit, to examine the influence of the type I interferon anti-viral response.

6. Determination of YAP1 protein expression in human head and neck cancer tissue

and normal tissue samples by immunohistochemistry (IHC) staining, in order to

assess the applicability of using YAP1 as a predictive biomarker of reovirus

treatment response.

156

5.3 RESULTS

5.3.1 Total YAP1 protein expression fluctuates after infection with reovirus in

SCCHN cell lines and in stable over-expressing YAP1 cell lines

In order to assess the effect of reovirus infection on YAP1 protein expression, we

measured total YAP1 levels pre- and post- infection with reovirus by flow cytometry

in PJ34, HN5 and PJ41 SCCHN cell lines, as well as in the EYFP-YAP1-clone 6,

Flag-YAP1-clone 2, and empty (EV)-clone 1 stable cell lines (Section 2.19). Cells

were permeabilised and stained with a total YAP1 primary antibody followed by an

Alexa Fluor® 546 secondary antibody. The cells were then analysed on the MACS

Quant® flow cytometer. Each cell line was also stained with secondary antibody

alone, which served as a negative control. The mean fluorescence intensity (MFI)

values for the negative control were subtracted from the positive MFI values in each

cell line.

PJ34, HN5 and PJ41 cell lines displayed low, medium and high YAP1 expression

respectively prior to infection with reovirus (0 hours post-infection (hpi)) (Figure 5.1

A), which correlated with the YAP1 expression pattern in these cells at the mRNA

level (Figure 3.4). Numerically, YAP1 expression in PJ41 cells was 13-fold and 47-

fold higher than in HN5 and PJ34 respectively at 0hpi, and was 4-fold higher in HN5

than in PJ34 cells. YAP1 expression in EYFP-YAP1-clone 6 was 11-fold higher than

in HN5, and was 8-fold higher in Flag-YAP1-clone 2 compared to empty (EV)-clone 1

cells at 0hpi (Figure 5.1 B). This supported our previous western blotting data

(Figures 4.7 and 4.8). Interestingly, EYFP-YAP1-clone 6 and Flag-YAP1-clone 2

over-expressed YAP1 almost to the levels of the PJ41 cell line (Figure 5.1 A and B).

There appeared to be a decrease in YAP1 expression after infection with reovirus in

most of the cell lines, most notably in PJ41, stable EYFP-YAP1 and stable Flag-YAP1.

However by 48 hours, the level of YAP1 was almost restored back to its natural levels

(Figure 5.1 A and B).

157

A B

Figure 5.1. The YAP1 protein fluctuates after infection with reovirus in the SCCHN and stable cell lines. The mean fluorescence intensity (MFI) of total

YAP1 was measured at 0, 8, 16, 24 and 48 hours post-infection (hpi) with reovirus at MOI 5 on the MACS Quant® flow cytometer in A. PJ41, HN5 and PJ34

SCCHN cell lines, and B. HN5 parental and stable empty-vector (EV), EYFP-YAP1 and Flag-YAP1 cell lines. At 0hpi, PJ34, HN5 and PJ41 cells showed low,

medium and high levels of YAP1 respectively which correlated with their susceptibilities to reovirus oncolysis. Results also confirmed the stable over-expression of

YAP1 in EYFP-YAP1 and Flag-YAP1 cell lines at 0hpi. There was a decrease in total YAP1 protein expression after infection with reovirus in the cell lines at 16

and 24hpi, but the level of YAP1 was almost restored back to its natural levels by 48hpi, particularly in PJ41, EYFP-YAP1 and Flag-YAP1 cell lines. The graphs

represent preliminary data generated from an independent experiment.

158

5.3.2 JAM-A protein expression did not correlate with the susceptibility of

SCCHN cell lines to reovirus oncolysis, and was not altered by stable-

over-expression of YAP1

To test whether the susceptibility of SCCHN cell lines to reovirus oncolysis correlated

with the expression of the main reovirus cellular receptor, the level of JAM-A was

determined by flow cytometry in PJ34, HN5 and PJ41 cell lines (Section 2.19). We

predicted to see greater JAM-A expression in the most sensitive cell line (PJ34), and

the least JAM-A expression in reovirus-resistant PJ41 cells. Cells were stained with a

JAM-A primary antibody followed by an Alexa Fluor® 546 secondary antibody, and

then analysed on the MACS Quant® flow cytometer. Each cell line was also stained

with secondary antibody alone, which served as a negative control. The mean

fluorescence intensity (MFI) values for the negative control were subtracted from the

positive MFI values in each cell line.

Figure 5.2 A demonstrates that cell surface expression of JAM-A was lowest in the

most sensitive cell line (PJ34), and the highest level of JAMA-A was observed in the

second most resistant cell line (HN5). There were no statistical differences in JAM-A

expression between the cell lines (un-paired t-test). Therefore, there was no clear

evidence that the level of JAM-A expression predicted for the differences in

susceptibility to reovirus oncolysis in these cell lines.

The level of JAM-A expression was also compared in HN5 parental cells, the stable

empty vector (EV) clone-1, and the stable Flag-YAP1-clone 2 and EYFP-YAP1-clone-

6 over-expressing cell lines. Cells were prepared and analysed in the same way as

described above. Figure 5.2 B shows that over-expression of YAP1 did not

significantly alter the level of JAM-A expression in the HN5 SCCHN cell line. There

were no statistical differences in JAM-A expression between the cell lines (un-paired

t-test). This suggested that YAP1-mediated restriction of reovirus oncolysis does not

occur at the cell surface, via the JAM-A receptor.

159

A

B

Figure 5.2. JAM-A expression does not correlate with the level of reovirus oncolysis in SCCHN

cell lines, and stable over-expression of YAP1 does not alter the level of JAM-A expression at the

cell surface. Cells were stained with a JAM-A primary antibody followed by an Alexa Fluor® 546

secondary antibody, before analysis on the MACS Quant® flow cytometer. A. There was no

correlation between JAM-A expression and the level of reovirus oncolysis in PJ34, HN5 and PJ41

SCCHN cell lines. B. There was no difference in JAM-A expression between stable-YAP1-over-

expressing cell lines (Flag-YAP1 and EYFP-YAP1) and the stable empty-vector (EV) clone or the

parental HN5 cell line. The graphs show the mean of two assay repeats and error bars represent SEM.

HN

5E

V

EY

FP

-YA

P1

Fla

g-Y

AP

1

0 .0

0 .2

0 .4

0 .6

0 .8

1 .0

1 .2

Me

an

Flu

ore

sc

en

ce

In

ten

sit

y (

MF

I)

PJ34

HN

5

PJ41

0 .0

0 .2

0 .4

0 .6

0 .8

1 .0

1 .2

Me

an

Flu

ore

sc

en

ce

In

ten

sit

y (

MF

I)

160

5.3.3 Reovirus protein can be detected in the cytoplasm of resistant and

sensitive SCCHN cell lines

In order to visualise reovirus in the SCCHN cell lines, and to verify whether stable

over-expression of YAP1 inhibits reoviral entry to the cell, an immunofluorescent

stain was performed in cells treated with media alone as a negative control, or in cells

infected with reovirus at MOI 5, 20 or 200 for 24 hours (Section 2.17). Cells were

permeabilised and stained with an anti-reovirus T3D antiserum. A secondary

antibody conjugated to a fluorescent dye was then used to visualise the reovirus

protein by confocal microscopy. PJ34 and PJ41 cell lines were probed with an Alexa

Fluor® 488 secondary antibody. Since the EYFP-tag fluoresces under the 488

wavelength of light, a different secondary antibody (Alexa Fluor® 546) was used to

compare reovirus infection between HN5 parental cells and the EYFP-YAP1 stable

cell line. A nuclear marker, TO-PRO-3, was used in all cell lines to help localise

reovirus in the cells. PJ34 and PJ41 cells were also stained with a cell membrane

marker, wheat germ agglutinin (WGA).

Reovirus was detected in the cytoplasm of all cell lines, which is where reovirus

replication is known to take place [99]. Reovirus-positive cells were detected to

similar levels at moderate MOI in sensitive PJ34 and resistant PJ41 cell lines. At high

MOI 200, PJ34 cells looked saturated with reovirus and there were fewer viable cells

than PJ41 (Figure 5.3). No co-localisation of reovirus and WGA was observed on the

cell surface, as this would have resulted in a yellow pigmentation. There appeared to

be equal numbers of HN5 parental and stable EYFP-YAP1 cells positive for reovirus

protein, although the staining intensity did appear slightly brighter in HN5 cells at

MOI 5 and 20 (Figure 5.4). However, it was difficult to definitively conclude this

due to the subjective nature of the assay, and a more quantitative method was required

to confirm any differences in intracellular reovirus protein production between these

cell lines. Although a cell membrane stain was not included for the HN5 and EYFP-

YAP1 cells, it was evident that reovirus protein was only localised inside the cells and

not on the cell membrane. Taken together with the JAM-A expression data, these

results suggested that the variation in susceptibility to reovirus-induced cell death in

PJ34 and PJ41 SCCHN cell lines, and the reovirus resistance accompanying YAP1

over-expression, does not seem to be cell surface receptor-mediated.

161

Figure 5.3. Reovirus infected both sensitive (PJ34) and resistant (PJ41) SCCHN cell lines at

different multiplicities of infection (MOI). Cells were treated with media alone (no reovirus) or

infected with reovirus at MOI 5, 20 or 200 for 24 hours. Resistant-PJ41 and sensitive-PJ34 cell lines

were permeabilised, stained with an anti-reovirus T3D antiserum and an Alexa Fluor® 488 secondary

antibody. Cells were then imaged using the same confocal microscope settings at ×40 magnification.

Reovirus protein was detected in both cell lines at all MOIs, as shown by the green staining. No green

staining was detected in the absence of reovirus infection. Wheat germ agglutinin (WGA) was used as

a cell membrane marker (red stain) and TO-PRO-3 was used to detect the cell nucleus (blue stain).

PJ34 PJ41

No reovirus

MOI 5

MOI 20

MOI 200

162

Figure 5.4. Reovirus can infect HN5 parental cells and the more resistant-EYFP-YAP1 stable cell

line at different multiplicities of infection (MOI). Cells were treated with media alone (no reovirus)

or infected with reovirus at MOI 5, 20 or 200 for 24 hours. EYFP-YAP1 and HN5 cell lines were

permeabilised, stained with an anti-reovirus T3D antiserum and an Alexa Fluor® 546 secondary

antibody. Cells were then imaged using the same confocal microscope settings at ×40 magnification.

We did not distinguish a difference in the number of cells infected between the two cell lines, but the

staining intensity appeared slightly brighter in HN5 than in EYFP-YAP1 cells. No red staining was

detected in the absence of reovirus infection. TO-PRO-3 was used to detect the cell nucleus (blue

stain).

HN5 EYFP-YAP1

No reovirus

MOI 5

MOI 20

MOI 200

163

5.3.4 The rate of intracellular reovirus protein production in PJ34, HN5 and

PJ41 SCCHN cell lines did not correlate with their reovirus IC50 values

As the differences in reovirus IC50 in the SCCHN cell lines are probably not due to

the prevention of reovirus entry at the cell membrane, we questioned whether it was

due to an obstruction of a step in the viral replication life cycle inside the host cell.

Intracellular reovirus yield was quantitatively measured in infected SCCHN cell lines.

Since increased reovirus replication has been shown to correlate with enhanced

reovirus oncolysis in various cancer cell lines [135, 142, 338-340], we predicted to

observe a similar trend in SCCHN cells.

Infectious reovirus titre was determined by one-step growth curve analysis via the

50% tissue culture infective dose (TCID50) assay (Section 2.20). PJ34, HN5 and PJ41

SCCHN cell lines were infected with reovirus at MOI 5 for 4, 20, 24, 48 or 72 hours.

Intracellular viral samples were prepared as described in Section 2.20.1, and used to

infect a monolayer of L929 mouse fibroblast cells. The cytopathic effect (CPE) in

each sample was determined using light microscopy after 3 days. In this experiment,

MOI 5 was used because reovirus was capable of infecting SCCHN cells at this

concentration, as displayed in Figure 5.3 and Figure 5.4. In addition, it was

important to use an MOI that would not saturate the resistance associated with the

PJ41 cell line or cause major cell death in the sensitive PJ34 cell line.

Figure 5.5 demonstrates viral growth in PJ34, HN5 and PJ41 SCCHN cell lines over

the 72 hour time-period. One cycle of reovirus replication takes approximately 18 to

24 hours in permissive cell lines [340]. Thus, by 20 hours, the viral titre in all cell

lines increased by at least 2-logs as newly synthesised virus proteins would have been

generated. Surprisingly, the viral titre between the cell lines, especially in HN5 and

PJ34, was not that different and lacked statistical significance (one-way ANOVA and

Tukey’s post-hoc test). The titres were not as well spread as what we would have

expected them to be (less than 1-log difference), considering their variable reovirus

IC50 values. The extensive over-lap in the viral growth curves of the cell lines led us

to believe that there was no distinct relationship between reovirus replication and

reovirus oncolysis.

164

Lysates were also collected from PJ34, HN5 and PJ41 cells infected with reovirus at

MOI 5 for 24, 48 or 72 hours. Total intracellular reovirus protein in the samples was

compared by western blotting (Section 2.14) by probing with an anti-reovirus T3D

antiserum.

We were unable to detect a clear band at 160kDa, which is the expected molecular

weight of the λ reovirus protein. The combination of MOPS running buffer and

Novex® 4-12% Bis-Tris gels used in these assays is apparently capable of detecting

proteins as large as 260kDa. Perhaps a lower percentage gel would have improved

the detection of the λ protein. Nevertheless, bands at approximately 80kDa and

40kDa corresponded to reovirus proteins µ and σ respectively. Figure 5.6 A shows

that there were subtle differences in total intracellular reovirus protein expression, but

it was not as variable as we predicted. The intensity of viral µ and σ bands in each

sample was quantified and normalised to their corresponding α-tubulin bands by

densitometry analysis (Section 2.14.4) (Figure 5.6 B). Total intracellular reovirus

protein production in these cells did not correlate with their reovirus IC50 values. For

example, there was little difference in the levels of µ and σ in PJ34 and HN5 at 24hpi

and 72hpi, and between HN5 and PJ41 cells at 48hpi. Admittedly, the α-tubulin

western blot showed some uneven loading, despite using the BCA protein assay for

equilibration (Section 2.14.1). However, normalising the reovirus protein bands to

their respective α-tubulin bands by densitometry showed that the western blotting data

generally supported the results of the TCID50 assay. This re-enforced our conclusion

that direct reovirus replication is probably not the predominant route of cell death

induced by reovirus infection in PJ34, HN5 and PJ41 SCCHN cell lines.

165

Figure 5.5. Infectious intracellular reovirus yield in PJ34, HN5 and PJ41 SCCHN cell lines did

not correlate with their reovirus IC50 values, as determined by the 50% tissue culture infective

dose (TCID50) assay. Intracellular infectious reovirus titre was evaluated in PJ34 (yellow triangles),

HN5 (orange squares) and PJ41 (red circles) SCCHN cell lines infected with reovirus at MOI 5 for 4,

20, 24, 48 or 72 hours. There was extensive over-lap in the viral growth curves of the cell lines. Viral

titre is shown on a log10 scale. The graph shows the mean of two assay repeats and error bars

represent SD.

0 4

20

24

48

72

1 0 3

1 0 4

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

T im e (h o u rs )

TC

ID5

0/m

L

P J 4 1

H N 5

P J 3 4

166

A

B

Figure 5.6. Total intracellular reovirus protein did not correlate with the susceptibility to

reovirus oncolysis in PJ34, HN5 and PJ41 SCCHN cell lines, as determined by western blotting.

Whole cell lysates were collected from PJ34, HN5 and PJ41 cells infected with reovirus at MOI 5 for

24, 48 or 72 hours. A. Reovirus proteins µ (80kDa band) and σ (40kDa band) were resolved on

Novex® 4-12% Bis-Tris gels and probed with an anti-reovirus T3D antiserum. The blots were imaged

at the same exposure enabling direct comparison of band intensity between the samples. B. The

intensity of viral µ and σ bands in PJ34 (yellow bars), HN5 (orange bars) and PJ41 (red bars) were

quantified and normalised to their corresponding α-tubulin bands at each time-point by densitometry

analysis.

µ

-tubulin

~ 80kDa

~40kDa

~50kDa

PJ34

HN

5

PJ41

PJ34

HN

5

PJ41

PJ34

HN

5

PJ41

72 hpi24 hpi 48 hpi

167

5.3.5 Intracellular reovirus protein production was hindered by stable over-

expression of YAP1

To test the specific effect of YAP1 over-expression on reovirus replication, we

measured the rate of intracellular reovirus yield in stably-transfected cell lines using

quantitative methodologies.

To begin with, infectious reovirus titre was determined by one-step growth curve

analysis via the TCID50 assay (Section 2.20) in the stable EV-clone-1, Flag-YAP1-

clone-2 and EYFP-YAP1-clone-6 cell lines, as well as in HN5 parental cells. Cells

were infected with reovirus at MOI 5 for 4, 20, 24, 48 or 72 hours. Intracellular viral

samples were prepared as described in Section 2.20.1. An MOI 5 was used to

evaluate the phenotypic restriction mediated by YAP1 over-expression, as high MOIs

were previously shown to dampen this effect (Figure 4.10 and Figure 4.11).

Figure 5.7 A demonstrates the rate of viral growth in stable EYFP-YAP1 cells and the

HN5 parental cell line. There was less intracellular reovirus produced over time in

EYFP-YAP1 cells than in HN5 cells, which was statistically (un-paired t-test) and

biologically significant. Likewise, the virus yield was significantly and consistently

lower, sometimes by more than 1-log, in stable Flag-YAP1 cells compared to stable

EV-control cells at all time-points tested (Figure 5.7 B), with clear separation in the

viral growth curves. This would partially explain the resistance to reovirus oncolysis

after YAP1 over-expression of HN5 cells.

Next, lysates were collected from HN5 parental cells, and stable EV-clone-1, Flag-

YAP1-clone-2 and EYFP-YAP1-clone-6 cell lines infected with reovirus at MOI 5 for

24, 48 or 72 hours. The production of total intracellular µ and σ reoviral proteins in

the samples was determined by western blotting (Section 2.14). The expression

levels of µ and σ was considerably lower in stable EYFP-YAP1 than in HN5 parental

cells at all time-points tested. There was some non-specific effect caused by the

stable EV-control cell line, although the expression of µ and σ was still lower in stable

Flag-YAP1 cells (Figure 5.8 and Figure 5.9). This supported the TCID50 data

displayed in Figure 5.7, which suggested that infectious viral titre correlated with

non-infectious viral titre.

168

Additionally, we measured total intracellular reovirus protein production in HN5

parental cells, and in EV-clone-1, Flag-YAP1-clone-2 and EYFP-YAP1-clone-6 stable

cell lines by flow cytometry (Section 2.19). Cells were infected with reovirus at MOI

5 for 16, 24, 48 and 72 hours, before being permeabilised and stained with an anti-

reovirus T3D antiserum, followed by an Alexa Fluor® 546 secondary antibody. The

samples were then analysed on the MACS Quant® flow cytometer. Each cell line

was also stained with secondary antibody alone, which served as a negative control.

The mean fluorescence intensity (MFI) values for the negative control were subtracted

from the positive MFI values in each cell line.

In comparison to HN5 parental or stable EV-cell lines, total intracellular reovirus

expression levels were substantially lower in stable EYFP-YAP1 and Flag-YAP1 cells

respectively, at all time-points tested (Figure 5.10 A and B). These results concurred

with the TCID50 data displayed in Figure 5.7 and with the western blots presented in

Figure 5.8 and Figure 5.9. There were some non-specific effects caused by the stable

EV-control cell line. It is also important to note that the viral titre in HN5 cells in

Figure 5.7 did vary slightly from the titre measured in Figure 5.5, which is likely due

to the fact that these experiments were performed at separate times in the project, and

with different cell passages. However, the viral titre could be compared between cell

lines within the same experiment.

169

A

B

Figure 5.7. The rate of infectious intracellular reovirus was hindered by stable over-expression

of YAP1 in the HN5 SCCHN cell line, as determined by the 50% tissue culture infective dose

(TCID50) assay. Intracellular infectious reovirus titre was compared in A. HN5 parental (blue

triangles) and stable EYFP-YAP1 (red circles) cells and B. HN5 parental (blue triangles), stable empty

vector (EV) cells (green triangles), or stable Flag-YAP1 (red circles) cells, infected with reovirus at

MOI 5 for 4, 20, 24, 48 or 72 hours. Viral titre was consistently lower in EYFP-YAP1 and Flag-YAP1

cell lines compared to HN5 and EV-control cells respectively. Viral titre is shown on a log10 scale.

**p<0.01, ***p<0.001 and ****p<0.0001 by un-paired t-test. The graphs show the mean of two assay

repeats and error bars represent SD.

0 4

20

24

48

72

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

1 0 1 0

T im e (h o u rs )

TC

ID5

0/m

L

H N 5 p a re n ta l


***

**** *****

**

0 4

20

24

48

72

1 0 4

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

1 0 1 0

T im e (h o u rs )

TC

ID5

0/m

L

H N 5 p a re n ta l

F L A G -Y A P 1 c lo n e 2

E V c lo n e 1

****

**** ******

****

170

A

B

Figure 5.8. The rate of total intracellular reovirus protein production was hindered by stable

over-expression of YAP1 in the HN5 SCCHN cell line, as determined by western blotting. Whole

cell lysates were collected from HN5 parental and stable EYFP-YAP1 cell lines infected with reovirus

at MOI 5 for 24, 48 or 72 hours. A. Reovirus proteins µ (80kDa band) and σ (40kDa band) were

resolved on Novex® 4-12% Bis-Tris gels and probed with an anti-reovirus T3D antiserum. The blots

were imaged at the same exposure enabling direct comparison of band intensity between the samples.

There was less µ and σ protein in stable EYFP-YAP1 than in HN5 cells at all time-points tested. B.

The intensity of viral µ and σ bands in HN5 (blue bars), EYFP-YAP1 (red bars) were quantified and

normalised to their corresponding α-tubulin bands at each time-point by densitometry analysis. The

relative density is shown on a log10 scale.

171

A

B


over-expression of YAP1 in the HN5 SCCHN cell line, as determined by western blotting. Whole

cell lysates were collected from HN5 parental, stable empty vector (EV) and stable Flag-YAP1 cell

lines infected with reovirus at MOI 5 for 24, 48 or 72 hours. A. Reovirus proteins µ (80kDa band) and

σ (40kDa band) were resolved on Novex® 4-12% Bis-Tris gels and probed with an anti-reovirus T3D

antiserum. The blots were imaged at the same exposure enabling direct comparison of band intensity

between the samples. There was less µ and σ protein in stable Flag-YAP1 than in stable EV and HN5

parental cells at all hours post-infection (hpi). B. The intensity of viral µ and σ bands in HN5 (blue

bars), stable EV (green bars) and stable Flag-YAP1 (red bars) cells were quantified and normalised to

their corresponding α-tubulin bands at each time-point by densitometry analysis. The relative density

is shown on a log10 scale.

172

A

B


over-expression of YAP1 in the HN5 SCCHN cell line, as determined by flow cytometry. The

mean fluorescence intensity (MFI) of total intracellular reovirus protein was measured at 16, 24, 48 and

72 hours post-infection (hpi) with reovirus at MOI 5 by the MACS Quant® flow cytometer in A. HN5

parental (blue bars) and EYFP-YAP1 (red bars) and B. HN5 parental (blue bars), stable empty-vector

(EV) (green bars) and stable Flag-YAP1 (red bars) cell lines. In comparison to HN5 parental or stable

EV-cell lines, total intracellular reovirus expression levels were substantially lower in stable EYFP-

YAP1 and Flag-YAP1 cells at all time-points tested. The graphs represent preliminary data generated

from an independent experiment.

173

5.3.6 Extracellular reovirus secretion was indistinguishable in the SCCHN cell

lines and was not hindered by stable over-expression of YAP1

Certain studies have suggested that reovirus oncolysis not only depends on effective

intracellular virus production, but also on the efficient release of progeny virus for

cell-to-cell spread. We therefore assessed whether there were any differences in

extracellular virus in infected SCCHN cell supernatants, and whether extracellular

virus secretion is affected by over-expression of YAP1 in HN5 cells.

Extracellular reovirus titre was determined by the TCID50 assay (Section 2.20) in the

most sensitive PJ34 and most resistant PJ41 SCCHN cell lines. In a separate

experiment, viral titre was also compared in stable EYFP-YAP1-clone-6 and HN5

parental cell lines. Cells were infected with reovirus at MOI 5 at various time-

intervals up to 72 hours. Extracellular viral samples were prepared as described in

Section 2.20.1 and used to infect a monolayer of L929 mouse fibroblast cells. The

cytopathic effect (CPE) in each sample was determined using light microscopy after 3

days.

Figure 5.11 A displays the rate of viral release in PJ34 and PJ41 cell lines. There

were statistical differences in extracellular virus yield between the cell lines at some

time-points post infection, albeit only having up to half a log difference. However,

overall there was much overlap in viral release over-time and there was no clear

distinction in extracellular viral yield at early or late times of infection. Thus, we

concluded there to be little difference in extracellular viral release between PJ34 and

PJ41 cell lines, despite their variable susceptibilities to reovirus oncolysis. Figure

5.11 B shows the rate of viral release in HN5 parental and stable EYFP-YAP1 cell

lines. Since more intracellular virus was detected in the parental HN5 cell line than

the stable EYFP-YAP1 cell line, we predicted to see a similar trend in the levels of

extracellular virus. Biologically, the difference in virus release between HN5 and

EYFP-YAP1 cells were not that relevant (less than half a log), despite being

statistically significant at some time points post-infection (un-paired t-test). If

anything, there was slightly more extracellular virus in stable EYFP-YAP1 cells, but

the viral growth curves were not well separated. This disproved our hypothesis, as

over-expression of YAP1 did not drastically alter extracellular secretion of reovirus.

174

A

B

Figure 5.11. There was little difference in extracellular reovirus secretion in PJ34 and PJ41

SCCHN cell supernatants, or in HN5 parental and stable YAP1 over-expressing cell

supernatants. Extracellular reovrus titre was compared in A. PJ34 (yellow triangles) and PJ41 (red

circles) cell supernatants or in B. HN5 parental (blue triangles) and stable EYFP-YAP1 (red circles)

cell supernatants, infected with reovirus at MOI 5 for up to 72 hours by the TCID50 assay. There was

extensive overlap in the viral growth curves with no clear separation between the cell lines, which

implied that there was no relationship between extracellular secretion of reovirus and reovirus

oncolysis. Viral titre is shown on a log10 scale. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001

by un-paired t-test. The graphs show the mean of two assay repeats and error bars represent SD.

0 1

4

8

16

24

48

72

1 0 4

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

T im e (h o u rs )

TC

ID5

0/m

L

P J 3 4

P J 4 1

***

****

****

*

***

0 1

4

8

24

48

72

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

1 0 1 0

T im e (h o u rs )

TC

ID5

0/m

L

H N 5 p a re n ta l


******

**

***

175

5.3.7 IFN-β secretion from SCCHN cell lines correlated with their respective

reovirus IC50 values, but was not consistently altered by over-expression

or knock-down of YAP1

Infection with many viruses, including reovirus, can stimulate a cascade of events that

can induce expression and secretion of type I IFN by the host cell, such as IFN-β.

Thus, the profile of IFN-β secretion before and after reovirus infection was assessed

in PJ34, HN5 and PJ41 SCCHN cell lines, which was compared to their respective

reovirus IC50 values. All SCCHN cell lines and human PBMCs (used as a positive

control) were treated with media (un-infected sample) or infected with reovirus at

various MOI for 24 hours. The supernatants were then harvested and used to quantify

the levels of secreted IFN-β by the Verikine™ Human IFN-β ELISA kit (Section

2.21). Figure 5.12 demonstrates the levels of IFN-β secreted by PJ34, HN5 and PJ41

SCCHN cell lines, with and without reovirus infection. As the MOI used to infect the

cells increased, so did the level of IFN-β secreted by the cells. PJ41 supernatants

contained the highest levels of IFN-β, HN5 displayed intermediate levels and PJ34

showed the lowest levels. This correlated with their sensitivities to reovirus-mediated

cell death. All infected SCCHN cells produced IFN-β to equal or greater levels than

the PBMC positive control. IFN-β was virtually un-detectable in un-infected cell

samples. A small amount of IFN-β was observed in un-infected PJ41 cell

supernatants, but the levels were below the limit of detection of the assay (<50

pg/mL) and it was therefore considered to be a negative value. IFN-β produced from

infected cells at an earlier time-point (twelve-hours) was also measured, which

showed a similar pattern in expression, but the levels were below the limit of

detection (data not shown).

Subsequently, in another assay, we determined whether the resistance associated with

over-expression of YAP1 is due to increased type I interferon signalling. IFN-β

secretion was evaluated in HN5 parental, and stable EV-clone-1, EYFP-YAP1-clone-6

and Flag-YAP1-clone-2 cell supernatants, pre- and post- infection with reovirus for 24

hours (Section 2.16.3), by the Verikine™ Human IFN-β ELISA kit (Section 2.21).

Non-infected and infected human PBMC cell supernatants were included as a positive

control for IFN-β secretion (Section 2.21.1). Figure 5.13 displays the level of IFN-β

produced by these cells. Compared to HN5 parental cells, there was no significant

176

alteration in IFN-β secretion in the stable EYFP-YAP1 cell line when infected with

reovirus at MOI 5 and MOI 100. In comparison to the stable EV-clone, stable Flag-

YAP1 cells produced almost 2-fold more IFN-β when infected with reovirus at MOI

100 (p<0.05 by un-paired t-test), but there was no difference in IFN-β expression at

MOI 5. Again, there was a positive dose-response relationship between reovirus MOI

and IFN-β secretion. All infected cell lines produced greater amounts of IFN-β than

the PBMC positive control. Small traces of IFN-β were detected in un-infected cell

line supernatants, but were regarded as negative values as they were below the limit of

detection.

In a separate experiment, we assessed whether the sensitivity associated with siRNA-

mediated knock-down of YAP1 in the PJ41 SCCHN cell line is a result of decreased

type I interferon signalling. PJ41 cells were transiently transfected with YAP1 siRNA

(ID: s20368) or negative control siRNA, or treated with Neo FX transfection agent

alone, or with media alone (Section 2.13.2). After 48 hours post-transfection, the

cells were infected with reovirus at MOI 5 or MOI 100, or treated with media alone

(no reovirus sample) for 24 hours (Section 2.13.3), before measuring IFN-β secretion

by the Verikine™ Human IFN-β ELISA kit (Section 2.21). Compared to the negative

control siRNA treated cells, there was no significant change in IFN-β production in

the cells transfected with YAP1 siRNA at MOI 5 or MOI 100 (Figure 5.14). A small

amount of IFN-β was produced in un-infected media alone cell supernatants, but it

was below the limit of detection and was therefore considered to be negative.

177

Figure 5.12. The levels of IFN-β secreted by PJ34, HN5 and PJ41 SCCHN cell lines after

infection with reovirus correlated with their sensitivities to reovirus oncolysis. SCCHN cell lines

and human PBMCs (used as a positive control for IFN-β) were treated with media (no reo, purple bars)

or infected with reovirus at MOI 5 (green bars), 10 (orange bars), 50 (blue bars) or 100 (yellow bars)

for 24 hours. The supernatants were then harvested and used to quantify the levels of secreted IFN-β

by the Verikine™ Human IFN-β ELISA kit. The limit of detection (50pg/ml) of the assay is displayed

on the graph as a solid line. The IFN-β levels are shown on a log10 scale and error bars represent the

SD from two independent experiments.

PJ34

HN

5

PJ41

PB

MC

1

1 0

1 0 0

1 0 0 0

1 0 0 0 0H

um

an

IF

N-B

(p

g/m

L) N O R E O

M O I 5

M O I 1 0

M O I 5 0

M O I 1 0 0

178

Figure 5.13. Stable over-expression of YAP1 in the HN5 SCCHN cell line did not consistently

increase the levels of IFN-β secretion after infection with reovirus. HN5 parental, stable empty-

vector (EV), EYFP-YAP1, Flag-YAP1 and human PBMC positive control cells were treated with media

(no reo, purple bars) or infected with reovirus at MOI 5 (green bars) or 100 (yellow bars) for 24 hours.

The supernatants were then harvested and used to quantify the levels of secreted IFN-β by the

Verikine™ Human IFN-β ELISA kit. Compared to HN5 parental cells, there was no significant

alteration in IFN-β secretion in EYFP-YAP1 cells infected with reovirus at either MOI. In comparison

to the EV-clone, Flag-YAP1 cells produced two-fold more IFN-β when infected with reovirus at MOI

100 (*p<0.05 by un-paired t-test), but there was no difference in IFN-β expression at MOI 5. The limit

of detection (50pg/ml) of the assay is displayed on the graph as a solid line. The IFN-β levels are

shown on a log10 scale and error bars represent the SD from two independent experiments.

HN

5 p

are

nta

lE

V

EY

FP

-YA

P-1

FL

AG

-YA

P-1

PB

MC

1

1 0

1 0 0

1 0 0 0H

um

an

IF

N-B

(p

g/m

L) N O R E O

M O I 5

M O I 1 0 0

*

179

Figure 5.14. siRNA-mediated knock-down of YAP1 in the PJ41 SCCHN cell line did not decrease

the levels of IFN-β secretion after infection with reovirus. PJ41 cells were transiently transfected

with YAP1 siRNA (ID: s20368) or negative control siRNA, or treated with Neo FX transfection agent

alone, or with media alone. After 48 hours post-transfection, the cells were treated with media alone

(no reo, purple bars), or infected with reovirus at MOI 5 (green bars) or MOI 100 (yellow bars) for 24

hours, before measuring IFN-β secretion by the Verikine™ Human IFN-β ELISA kit. There was no

significant difference in IFN-β production between negative control siRNA treated cells or the cells

transfected with YAP1 siRNA at either MOI. The limit of detection (50pg/ml) of the assay is displayed

on the graph as a solid line. The IFN-β levels are shown on a log10 scale and error bars represent the

SD from two independent experiments.

Med

ia o

nly

YA

P1 s

iRN

A (

s20368)

Neg

at i

ve s

iRN

A

Neo

FX

on

ly

PB

MC

1

1 0

1 0 0

1 0 0 0H

um

an

IF

N-B

(p

g/m

L) N O R E O

M O I 5

M O I 1 0 0

180

5.3.8 Detection of the YAP1 protein in SCCHN tissue and normal tissue

Our results so far suggest that YAP1 has the potential to be used as a biomarker to

predict the anti-tumour effects of reovirus in clinical applications of SCCHN. The

likely method of detection of the YAP1 protein would be staining of tumour tissue

samples directly resected from SCCHN patients. Therefore, we determined whether

YAP1 could be detected not only in SCCHN cell lines, but also in SCCHN tissue, and

whether there was any difference in YAP1 expression compared to in normal tissues.

First, the enzymatic immunohistochemistry (IHC) staining protocol (Section 2.22)

was optimised in prostate cancer (PCa) tissue, which was recommended as the

positive control by the supplier of the YAP1 primary antibody (Abcam, UK). The

Human Protein Atlas [341] and published research [222] have also shown positive

nuclear and cytoplasmic YAP1 staining in different PCa tissues. Sections of the

tissue were stained using different dilutions of the primary antibody. Figure 5.15

shows that the PCa tissue was positive for the YAP1 protein (cytoplasmic and nuclear

brown staining), and the optimal primary antibody dilution was 1:400. The procedure

was also performed in the absence of the primary antibody as a negative control,

which was completely negative and only displayed blue coloration as a result of

counter staining with haematoxylin.

The enzymatic IHC staining procedure (Section 2.22) for the detection of YAP1 was

then performed on a head and neck cancer tissue microarray (US Biomax, cat:

HN803a). A system was utilised to score the tissues as 0, +1, +2 or +3, according to

the intensity of brown coloration. Upon inspection of the head and neck cancer cores,

it was clear that some did not contain any evidence of tumour. Therefore, these were

excluded from the analysis as they would not represent a true evaluation for YAP1

expression in head and neck cancer. Out of a total of 64 carcinoma of the head and

neck tissue cores, 8 (13%) stained positive for the YAP1 protein. The cellular

localisation of YAP1 in these tissues varied, as 4 cores displayed cytoplasmic, 1

showed nuclear, and 3 exhibited both cytoplasmic and nuclear staining. The array

also contained 10 normal tissues derived from tongue or pharynx, and 1 normal

adjacent tissue derived from tongue. All stained negative for YAP1. Examples of 0,

+1, +2 and +3 IHC YAP1 intensity staining for both normal tissue and tumour tissue

are shown in Figure 5.16. Demographic details for the tissues on the HN803a array

181

are shown in Table 5.1, including the IHC YAP1 intensity scores and YAP1 cellular

localisation.

Enzymatic IHC (Section 2.22) was then used to stain a multiple organ normal tissue

array (US Biomax, cat: FDA999c) for the YAP1 protein. Out of a total of 99 tissue

cores on the array, 9 were normal head and neck tissues. All 9 normal head and neck

tissues stained negative for YAP1. All but two of the other remaining normal tissues

on the array stained negative for YAP1. The two normal tissues that stained positive

for YAP1 originated from kidney and skin. Both of these tissues showed +1 intensity

and cytoplasmic YAP1 staining (data not shown). Demographic details for the tissues

on the FDA999C array are displayed in Table 5.2, including the IHC YAP1 staining

intensity scores and cellular localisation. All tissues were checked and confirmed by

Dr Silvana Di Palma, a consultant pathologist at The Royal Surrey Hospital.

YAP1 positive or negative staining of the tissue cores was compared by using the

Chi-squared (χ2) statistical test, which confirmed that there was a significant

difference between the head and neck carcinomas and normal tissues from all types of

organ (p=0.0035) (Table 5.3). Comparison of the head and neck carcinomas and the

normal head and neck tissues alone however, did not quite reach statistical

significance (p=0.097) (Table 5.3). There were no significant differences between

tumour grade, age of the patient, sex of the patient or tumour stage (Table 5.3). IHC

YAP1 staining intensity scores of the positive head and neck cancer tissues were then

compared against the cellular localisation of YAP1, tumour grade, age of the patient

and tumour stage. Although the sample numbers were small, there were no apparent

statistical differences in these parameters and YAP1 intensity score (Table 5.4). No

follow-up or treatment information was provided with the HN803a head and neck

cancer array.

182

Figure 5.15. Optimisation of the enzymatic immunohistochemistry (IHC) staining protocol in

prostate cancer tissue for the detection of the YAP1 protein. The enzymatic IHC staining protocol

was performed on prostate cancer tissue sections as a positive control for the detection of the YAP1

protein. In the image on the right, the brown coloration represented positive YAP1 staining in the

presence of the YAP1 antibody at an optimal dilution of 1:400. As a negative control, the procedure

was also performed in the absence of the primary antibody (left image), which showed no brown

coloration. Images were photographed at ×20 magnification.

183

Figure 5.16. Examples of YAP1 protein expression from the FDA999c normal tissue array and

the HN803a head and neck cancer tissue array, using enzymatic immunohistochemistry (IHC). Enzymatic IHC examples of A. 0 intensity staining of normal tongue B. 0 intensity staining of

squamous cell carcinoma (SCC) of larynx C. +1 staining of SCC of larynx D. +2 staining of SCC of

submaxilla and E. +3 staining of SCC of laryngeal pharynx. The staining intensity score is indicated as

0, +1, +2, or +3. YAP1 positive staining in the cytoplasm is shown by the yellow arrows, and nuclear

staining is shown by the red arrows. Images were taken at ×10, ×20 and ×40 magnification.

184

Table 5.1. Demographic data from the HN803a head and neck cancer tissue array, with details of IHC YAP1 intensity scoring and cellular

localisation. The array contained a total of 64 cores, each representing a single case. Tissues highlighted in orange were positive for YAP1. Normal or

normal adjacent head and neck tissue (NAT) on the array are highlighted in green.

186

Table 5.2. Demographic data from the FDA999c multiple organ normal tissue array, with details

of IHC YAP1 intensity scoring and cellular localisation. The array contained a total of 99 cores,

each representing a single case. Tissues highlighted in orange were positive for YAP1. Normal or

normal adjacent head and neck tissue (NAT) on the array are highlighted in green.

188

Table 5.3. Statistical comparisons of YAP1 protein expression in the tissue sections by the Chi-

squared (χ2) statistical test. There was a significant difference in YAP positive (+ive) or YAP1

negative (-ive) staining between head and neck (H&N) carcinoma tissues and all (n=110) of the normal

tissues (χ2=8.523, p=0.0035, as highlighted in red). However, there was no difference between H&N

carcinoma tissues and the normal H&N tissues only (n=20), or between tumour grade, age of the

patient, sex of the patient or tumour stage.

n (total) n (YAP1 -ive) n (YAP1 +ive) χ2 p

Normal (all H&N) 20 20 0

Carcinomas of the H&N 64 56 8

n (total) n (YAP1 -ive) n (YAP1 +ive) χ2 p

Normal (all tissue types) 110 108 2

Carcinomas of the H&N 64 56 8

Tumour Grade (Carcinomas of the H&N) n (total) n (YAP1 -ive) n (YAP1 +ive) χ2 p

Grade 1 11 11 0

Grade 2 34 28 6

Grade 3 16 14 2

Age (Carcinomas of the H&N) n (total) n (YAP1 -ive) n (YAP1 +ive) χ2 p

30-39 1 1 0

40-49 15 10 5

50-59 20 19 1

60-69 16 15 1

70-79 11 10 1

≥ 80 1 1 0

Sex (Carcinomas of the H&N) n (total) n (YAP1 -ive) n (YAP1 +ive) χ2 p

Male 54 46 8

Female 10 10 0

Tumour Stage (Carcinomas of the H&N) n (total) n (YAP1 -ive) n (YAP1 +ive) χ2 p

Stage I 5 5 0

Stage II 20 18 2

Stage III 18 17 1

Stage IV 13 11 2

7.955 0.16

1.693 0.19

1.437 0.70

2.763 0.097

8.523 0.0035

2.279 0.32

189

Table 5.4. Statistical comparison of immunohistochemistry (IHC) YAP1 staining intensity in the

head and neck (H&N) carcinoma tissue sections by the Chi-squared (χ2) statistical test. There

were no significant differences in the IHC YAP1 staining intensity scores of the positive H&N

carcinoma tissues and cellular localisation of YAP1, tumour grade, age of the patient or tumour stage.

It was impossible to calculate any statistical differences for sex of the patients, as no female tissue

cores were positive for YAP1.

YAP1 Cellular Localisation (Carcinomas of the H&N) n (total) n (1) n (2) n (3) χ2 p

Cytoplasmic 4 3 1 0

Nuclear 1 1 0 0

Both 3 0 1 2

Tumour Grade (Carcinomas of the H&N) n (total) n (1) n (2) n (3) χ2 p

Grade 1 0 0 0 0

Grade 2 6 3 2 1

Grade 3 2 1 0 1

Age (Carcinomas of the H&N) n (total) n (1) n (2) n (3) χ2 p

40-49 5 3 1 1

50-59 1 0 0 1

60-69 1 1 0 0

70-79 1 0 1 0

Sex (Carcinomas of the H&N) n (total) n (1) n (2) n (3) χ2 p

Male 8 4 2 2

Female 0 0 0 0

Tumour Stage (Carcinomas of the H&N) n (total) n (1) n (2) n (3) χ2 p

Stage I 0 0 0 0

Stage II 2 1 1 0

Stage III 1 0 1 0

Stage IV 2 1 0 1

n/a n/a

IHC YAP1 Staining intensity score

3.750 0.44


7.200 0.30



6.167 0.19


1.333 0.51

190

5.4 DISCUSSION

The work illustrated in Chapters 3 and 4 suggested that the expression level of the

YAP1 protein is an important host cell determinant of sensitivity to reovirus oncolysis

in SCCHN cell lines. Consequently, the objective of this chapter was to investigate

the mechanism by which YAP1 mediates its effects on reovirus-induced cell death.

To assess this, we performed several experiments on the PJ34, HN5 and PJ41 SCCHN

cell lines. The stable cell lines generated from HN5 were also included due to their

reliable in vitro behaviour, and to allow us to study the specific effects of YAP1 over-

expression. The ability of reovirus to enter the cells was determined by measurement

of cell surface expression of JAM-A by flow cytometry. Reovirus replication

efficiency in the cell lines was ascertained by TCID50 one-step growth curves and

western blotting. Influence of type I interferon anti-viral responses was assessed by

quantifying cellular IFN-β secretion. Finally, in order to gain an insight into the

applicability of using YAP1 as a predictive biomarker of reovirus therapy, YAP1

protein expression in head and neck carcinoma tissue was compared to that in normal

tissue by IHC staining.

Total YAP1 protein expression was quantitatively measured pre- and post- infection

with reovirus by flow cytometry in the SCCHN cell lines. The results supported our

earlier RT-qPCR data, and confirmed that PJ34, HN5 and PJ41 cells displayed low,

medium and high levels of the YAP1 protein respectively, prior to infection with

reovirus. It was interesting to discover that both the EYFP-YAP1-clone 6 and Flag-

YAP1-clone 2 stable cell lines over-expressed YAP1 to similar levels as the PJ41 cell

line. This suggested that artificial over-expression of YAP1 in HN5 cells does not

behave in the same way as PJ41 cells, which naturally express high levels of YAP1,

as PJ41 cells (IC50 MOI 572.9 at 24hpi) were still more resistant to reovirus oncolysis

than stable EYFP-YAP1 (IC50 MOI 187.5 at 24hpi) and Flag-YAP1 cells (IC50 MOI

391.9 at 24hpi). This implied that YAP1 is a factor that contributes to the degree of

reovirus oncolysis, but other proteins may play a part in this too in different SCCHN

cell lines. We know that YAP1 is predominantly cytoplasmic in PJ41 cells, although

there was some nuclear YAP1 detected too (Section 4.3.7). However, the cellular

localisation of plasmid-mediated YAP1 over-expression in HN5 is un-known.

Different proportions of cytoplasmic and nuclear YAP1 may impact on its function

191

[251, 253, 275, 342], which may account for differences in the susceptibility to

reovirus oncolysis in PJ41, EYFP-YAP1 and Flag-YAP1. For future reference, it

would be interesting to perform live cell imaging to establish YAP1 localisation pre-

and post-infection with reovirus in these cell lines. Post-reovirus infection, it was

surprising to observe a marked decrease in the levels of YAP1, which then returned

back to its pre-infection levels at later times of infection. This pattern of expression

was most evident in the cells that expressed the highest levels of YAP1 and were the

most resistant to reovirus-induced cell death, i.e. PJ41, stable EYFP-YAP1 and stable

Flag-YAP1 cell lines. It is un-clear why a drop in the level of YAP1 occurs after

incubation with reovirus, as this implies that YAP1 is not a protein that is induced in

response to reovirus infection. This is unlike other known anti-reoviral proteins such

as interferon-beta (IFN-β) [331, 332], secretogranin 2 (SCG2) [343] and interferon-

inducible transmembrane protein 3 (IFITM3) [230], whose expression and secretion

are stimulated after reoviral infection. However, inhibition of host cell DNA

synthesis is one of the earliest cytopathic effects observed after reovirus infection in

cultured cells [99, 344]. Reovirus infection can also cause inhibition of cellular RNA

and/or protein synthesis [99]. Therefore, perhaps YAP1 is one such protein that the

virus naturally suppresses in cells that express a certain level of YAP1. Down-

regulation of cellular FLICE inhibitory protein (cFLIP) and Akt by reovirus infection

sensitised human ovarian and gastric cancer cell lines to TRAIL-induced apoptosis

[345, 346]. It is possible that YAP1 has an anti-apoptotic function, as mentioned later

in this discussion. If YAP1 is only partially down-regulated by reovirus and some

expression still remains, then this may be sufficient for the cell to survive infection

and eventually, more YAP1 is synthesised by the cell, which my explain the increase

in YAP1 at later time-points. From our experiment, there is strong evidence to

suggest that reovirus infection does not affect the regulation of the cytomegalovirus

(CMV) promoter in the stable-plasmid-transfected cell lines, because the same pattern

of YAP1 expression was observed in the un-transfected, PJ41 cell line.

Attachment of reovirus to the host cell is a multistep process that is initiated by the

reovirus 1 protein binding to sialic acid on the cell surface membrane with low

affinity [110]. The head domain of 1 then makes contact with the JAM-A cellular

receptor with high affinity before the virions become internalised [110]. We therefore

determined whether the method of reovirus entry into SCCHN cells through the main

192

reovirus cellular receptor, JAM-A, would predict for their sensitivity to reovirus

oncolysis. It was hypothesised that the expression levels of JAM-A would be highest

in PJ34 and lowest in PJ41 cells. However, JAM-A expression did not correlate with

the susceptibility of reovirus oncolysis in PJ34, HN5 or PJ41, as its expression was

lowest in the most sensitive cell line (PJ34), and highest in the second most resistant

cell line (HN5). This was in agreement with the results published by Twigger et al,

who found that the levels of JAM-A measured in HN5 and three other SCCHN cell

lines (HN3, Cal27 and SIHN-5B) did not show a relationship with their corresponding

reovirus IC50 values [145]. JAM-A is usually expressed in endothelial and epithelial

cells of various tissue types. Many cancers express high levels of JAM-A to aid their

proliferation and metastasis [347]. However, other cancers such as gliomas and

melanomas express limited JAM-A [347]. Expression of the JAM-A receptor was not

a major determinant of reovirus sensitivity in glioblastoma stem-like cells or

colorectal liver metastasis [348, 349], or in a number of different cancer cell types

including breast, lung, prostate and bladder cancer cell lines [144]. In fact, reovirus

can enter the cell via a different route independently of JAM-A cell-surface binding.

Removal of the outer capsid protein 3 and cleavage of µ1 to µ1C can occur outside

of the cell to generate intermediate subviral particles (ISVPs), which can directly

penetrate the cell membrane [142]. Therefore, our findings that JAM-A expression

did not correlate with the susceptibility of SCCHN cell lines to reovirus oncolysis, are

not surprising. Intriguingly, over-expression of YAP1 did not change the level of

JAM-A expression compared to the parental HN5 or the stable EV-control cell lines.

It was therefore deduced that YAP1-mediated restriction of reovirus oncolysis does

not primarily occur at the cell surface, via the JAM-A receptor. Further supporting

this result, reovirus protein could be visually detected in the cells by

immunofluorescence staining and confocal microscopy, even in the more resistant cell

lines (PJ41 and stable EYFP-YAP1) at relatively low MOI. Hence, it was believed

that the variation in the susceptibility to reovirus-induced cell death was more likely

due to a step in the viral replication life cycle being affected inside the host cell.

The rate of infectious intracellular reovirus yield was quantified by the TCID50 assay

in PJ34, HN5 and PJ41 SCCHN cell lines. The intracellular titre was not as well

spread as we originally predicted, considering the variation in reovirus IC50 values in

these cell lines. There was extensive overlap in the intracellular viral growth curves,

193

and the western blotting data highlighted that there was little difference in total

reovirus proteins (infectious ISVPs and non-infectious viral cores) produced in these

cells. For example, the levels of µ and σ reovirus proteins in PJ34 and HN5 at 24hpi

and 72hpi were very similar, as were the levels in HN5 and PJ41 at 48hpi.

Correspondingly, there was little difference in extracellular viral secretion in PJ34,

HN5 and PJ41 cells. Overall, the lack of a clear association between viral replication

and reovirus oncolysis suggested that the mode of killing in these cells is not

influenced by direct reovirus replication, and that other pathways determine the cells’

fate after infection. Likewise, HN5, HN3, Cal27 and SIHN-5B SCCHN cell lines all

showed the same level of reovirus replication, despite there being a 3-log spread in

their reovirus IC50 values [145]. Thus, it is not unexpected to see that PJ34, HN5,

PJ41 SCCHN cells behave in a similar manner in our experiments. Reovirus

replication did not correlate closely with cell sensitivity to reovirus-induced cell death

in the MeWo melanoma cell line [335], or in six different colorectal cancer cell lines

[350]. This phenomenon is further substantiated by the fact that non-replicative ultra-

violet (UV)-inactivated reovirus is still able to cause significant (though reduced

compared to non-UV irradiated counterparts) cell death in melanoma cell lines,

possibly because the virus is unstable and exposes its dsRNA, which is recognised by

cell sensors [335]. More specifically, reovirus-induced-apoptosis is not always

intimately linked with reovirus replication, as demonstrated by numerous different

studies [243, 321, 351, 352]. All of this research largely agrees with our findings. On

the contrary, several lines of evidence suggest that activated Ras signalling inhibits

the anti-viral activity of PKR to allow for increased reovirus replication, which

correlated with enhanced reovirus oncolysis in various cancer cell lines [135, 338].

Activated Ras can enhance the disassembly of the virus particle within the endosomal

compartment, the infectivity of progeny virions, and the release of progeny virions

[142, 339, 340]. The fact that our data opposed these results is not unforeseen, as Ras

status was not a key contributor to reovirus oncolysis in PJ34, HN5 and PJ41 SCCHN

cell lines [145], as mentioned in Sections 1.3.3.2 and 3.1.

We did not observe any major differences in extracellular reovirus yield in the

supernatants of infected stable EYFP-YAP1 and HN5 parental cell lines. This implied

that over-expression of YAP1 does not interfere with secretion of reovirus. In

contrast, there was less intracellular reovirus protein produced over time in EYFP-

194

YAP1 cells than in HN5 cells, which was statistically and biologically significant over

the seventy-two hours infection period. Similarly, virus yield was significantly and

consistently lower in stable Flag-YAP1 cells compared to stable EV-control cells at all

time-points tested, with a clear separation in the viral growth curves, which was

confirmed using three different methodologies. Thus, YAP1 over-expression may

directly obstruct a step in virus replication in the host cell, but not with its secretion,

which would in part explain the increased resistance to reovirus oncolysis.

Considering that there was no difference in intracellular reovirus titre measured in

PJ34, HN5 and PJ41 cell lines (whose expression levels of YAP1 also correlated with

their respective reovirus IC50 values), suggested that forced over-expression of YAP1

mediates its resistance to reovirus-induced cell killing differently to cells that

endogenously express YAP1. We do not fully understand why this is so, but again, it

could be due to differences in the cytoplasmic and nuclear expression levels of YAP1.

Perhaps more nuclear YAP1 would enhance cell proliferation to slow reovirus

replication, whereas more cytoplasmic YAP1 would reduce cell proliferation to

increase the rate of reovirus replication. Alternatively, other signalling pathways may

be activated after reovirus infection in PJ34, HN5 and PJ41 cells, which may be

dampening the specific effect of YAP1 on intracellular reovirus replication. Perhaps

forced-YAP1 over-expression is able to surpass the molecular signalling repertoire

that would normally contribute to reovirus-induced oncolysis. Our knock-down and

over-expression studies in Chapters 3 and 4 implied that YAP1 is an important

determinant of reovirus oncolysis in PJ41 and PJ34 SCCHN cells. Because we only

measured reovirus yield after over-expression of YAP1 in HN5 cells, we do not know

if reovirus replication would also be affected by knock-down of YAP1 in PJ41, or

transient over-expression of YAP1 in PJ34, or whether the effect is cell line specific.

As interferon stimulated genes (ISGs) such as IFITM3 have been shown to restrict

reovirus replication comparably to YAP1 over-expression [230], it seemed logical to

ascertain whether YAP1 functions as part of the type I interferon pathway. The levels

of IFN-β secretion were raised by two-fold in stable Flag-YAP1 cells compared to the

stable EV-control cell line infected with reovirus at MOI 100, but there was no

difference in IFN-β expression in these cell lines at MOI 5. Moreover, we found no

difference in IFN-β secretion between stable EYFP-YAP1 and HN5 parental cells.

When IFN-β was measured in these infected cells at an earlier time-point (12 hpi), no

195

difference in expression was observed (data not shown), despite the levels being

below the limit of detection of the assay. As the increase in IFN-β in the Flag-YAP1

cell line was not supported by the other stable YAP1 over-expressed cell line (EYFP-

YAP1), it was concluded that the resistance to reovirus oncolysis and the decreased

intracellular reovirus yield associated with over-expression of YAP1, acts

independently of, and is not linked to, the type I interferon anti-viral response. In

support of this, the reovirus sensitivity associated with siRNA-mediated knock-down

of YAP1 in the PJ41 cell line was not due to type 1 interferon signalling, as the levels

of IFN-β were not statistically different in negative control siRNA or YAP1 siRNA

transfected cells. On the other hand, IFN-β secretion correlated with the susceptibility

to reovirus oncolysis in PJ34, HN5 and PJ41 SCCHN cell lines. Viral supernatants of

sensitive PJ34 contained the lowest amount of IFN-β, HN5 had intermediate levels,

and resistant PJ41 cells expressed the highest levels. This proposed that type I

interferon signalling has an influence on the susceptibility to reovirus-induced cell

death in these cell lines. This did not agree with the results of Twigger et al, as they

found no correlation with reovirus sensitivity and IFN-β secretion in four

representative SCCHN cell lines (HN5, HN3, Cal27 and SIHN-5B) [145]. However,

efficient interferon signalling would prevent translation of viral proteins. As we did

not detect any difference in intracellular or extracellular reovirus protein production in

PJ34, HN5 and PJ41, this suggested that the effects of type I interferon are somewhat

overridden. It has been documented that reoviruses have evolved specific

mechanisms to evade the type I interferon anti-viral response. Firstly, the reovirus µ2

protein, which is involved in viral RNA synthesis, can provoke an unusual build-up of

the transcription factor IRF9 in the nucleus and repress ISG expression, possibly by

disrupting IRF9-STAT2-STAT1 interactions that are needed for IFN signalling [332].

Secondly, the reovirus σ3 protein can bind to dsRNA and inhibit activation of the

host-cell anti-viral protein PKR [332, 353]. In addition, 65-70% of tumours are

unable to produce or respond to type I interferon in order to escape anti-proliferative

or pro-death signals; an aberration that many oncolytic viruses take advantage of

[354], including recombinant vaccinia virus [355], vesicular stomatitis virus [356],

and reovirus [334]. Ras transformed cells have been shown to be incapable of

producing or responding to IFN-β by blocking signalling from RIG1, thus making the

cells unable to recognise viral RNAs [333]. Therefore, even though the SCCHN cell

lines used in this study were able to secrete IFN-β to levels that correlated with their

196

reovirus IC50 values, they may contain defects (presumably not in Ras signalling

[145]) in anti-viral innate immune responses that render them non-responsive to type I

interferon, allowing reovirus replication to proceed.

So far, we have not found a common mechanism linking YAP1 expression and the

susceptibility to reovirus oncolysis in all of the SCCHN cell lines. We intend to

perform gene expression profiling and microarray hybridisation on the stable cell

lines, infected with and without reovirus, in order to help uncover the signalling

pathways involved in this process. Since reovirus-induced apoptosis is not always

dependent on reovirus replication, perhaps YAP1 mediates its effects through

suppression of reovirus-induced apoptosis. Indeed, YAP has been shown to be a

regulator of apoptosis. When expressed in the nucleus, YAP can transcriptionally up-

regulate the expression of the anti-apoptotic factors Bcl-xL (encoded by the BCL2L1

gene and a member of the Bcl-2 family), and Survivin (encoded by the BIRC5 gene

and a member of the inhibitor of apoptosis (IAP) family) in certain cancer cell lines

[300]. Lin et al demonstrated that YAP enhanced the expression of Bcl-xL to

promote survival and resistance to RAF and MEK inhibitors in tumours harbouring

BRAF and RAS mutations [357]. This implied that YAP and RAF-MEK signalling

work together to regulate Bcl-xL [357]. Even though baseline levels of activated

GTP-loaded Ras in PJ34, HN5 and PJ41 SCCHN cell lines did not correlate with their

susceptibility to reovirus-induced cell death [145], perhaps activating Ras mutations

are not always essential for YAP1 to restrict reovirus-induced apoptosis in these cells.

Importantly, mitochondrial apoptotic signalling is involved in reovirus-induced

apoptosis. Over-expression of Bcl-2, an anti-apoptotic protein similar to Bcl-xL, can

inhibit reovirus-induced apoptosis by preventing the release of Smac and cytochrome

c [323], and by inhibiting the proteolytic cleavage and degradation of cellular IAPs,

including Survivin, cIAP1 (encoded by the BIRC2 gene) and XIAP (encoded by the

XIAP gene) [242] [243]. We therefore hypothesise that a certain level of YAP1

expression might prevent apoptosis induced by oncolytic reovirus, by promoting

expression of Bcl-xL, Bcl-2, Survivin or certain IAPs, which may inhibit downstream

mitochondrial apoptotic signalling, to aid survival in SCCHN cell lines. In HN5,

HN3, Cal27 and SIHN-5B SCCHN cell lines, reovirus-induced cell death was not

prevented by a pan-caspase inhibitor (z-VAD-FMK) and did not involve caspase 3

activation, suggesting that reovirus killing in these cells is non-apoptotic [145].

197

Despite this finding, it would be of importance to verify whether YAP1 hinders

reovirus-induced apoptosis in PJ34, HN5 and PJ41 SCCHN cell lines. Unfortunately,

there was insufficient time left in the project to fully test this theory, but it is

something to consider for future investigation.

Our results strongly suggest that host-cell expression levels of YAP1 influence the

susceptibility of SCCHN cells to reovirus-induced cytotoxicity, but that other un-

known factors are also clearly inter-linked with YAP1 signalling. There is

considerable inter- and possible intra- tumour heterogeneity in SCCHN, and the

signalling pathways in these tumour cells are extensively interconnected with high

levels of redundancy. This makes treatment of SCCHN and the identification of

predictive biomarkers to oncolytic reovirus therapy challenging. Multiple factors may

cooperate with YAP1 to determine the cells’ fate after reovirus infection, and the

importance of these factors may be variable in different SCCHN cell lines. Despite

not being able to fully understand the mechanism of how YAP1 expression influences

reovirus oncolysis, this work provides reason to further test YAP1 as a biomarker of

reovirus treatment response in SCCHN patient tumours. Our preliminary work of

YAP1 protein expression in tissue sections showed that a small population of head

and neck carcinoma tissues express YAP1, but different types of normal tissues do not

generally express it. We did not find any relationship between immunohistochemistry

YAP1 staining status and tumour grade, age of the patient, sex of the patient or

tumour stage. However, larger numbers of tissue samples would be needed to make

more comprehensive conclusions. Xu et al reported YAP1 to be a prognostic marker

for overall survival and disease-free survival in hepatocellular carcinoma patients

[283]. It would be interesting to analyse YAP1 and reovirus protein expression in

tumour tissue obtained from a Reolysin® clinical study, and to see if there is a

correlation with reovirus resistance, or survival outcome.

It could be argued that the work performed in this study has limitations, as in vitro

sensitivity to viral replication and oncolysis sometimes does not match the sensitivity

of a tumour type in vivo. Studying factors in cancer cells that influence reovirus

oncolysis in culture does not fully take into consideration the requirement of the

immune system in response to viral infection. Carrying out experiments on a greater

number of SCCHN cell lines would further substantiate our conclusions.

198

5.5 CONCLUSION

This project has shown that the efficiency of reovirus-induced cancer cell death

depends on the expression level of YAP1 in SCCHN cells, implying that YAP1 may

be an appropriate predictive biomarker of reovirus oncolysis. In this chapter, we

were unable to define the exact mechanism behind this, which is likely to be

complex, given the heterogenic nature of SCCHN tumour cells. However, we were

able to dismiss the possibility that reovirus entry is restricted at the cell surface via

the JAM-A receptor. Our data suggested that the cell lines are somewhat un-

responsive to, or reovirus itself evades, the effects of the type I interferon response,

as there was no difference in the rate of reovirus production or release in the cells.

Forced-over-expression of YAP1 did not affect type I interferon secretion, but did

restrict reovirus yield somewhat in the HN5 cell line for reasons we cannot fully

explain. YAP1 can transcriptionally up-regulate the expression of anti-apoptotic

factors that are also known to prevent reovirus-induced apoptosis. Therefore, we

predict that a certain level of YAP1 expression in SCCHN cell lines might prevent

apoptosis induced by reovirus infection through the promotion of mitochondrial

apoptotic signalling components. This would be an interesting lead for the

continuation of this project. As a final point, our data provides novel information

that may aid the clinical application of reovirus in patients with SCCHN.

199

CHAPTER 6

COMBINING REOVIRUS WITH

CHEMOTHERAPEUTIC TAXANE DRUGS IN PCa

CELL LINES

200

6. COMBINING REOVIRUS WITH CHEMOTHERAPEUTIC TAXANE DRUGS

IN PCa CELL LINES

6.1 INTRODUCTION

The full potential of using oncolytic viruses to treat cancer in the clinical setting is

likely to be achieved when combined with other treatment modalities. Many studies

have shown that combining reovirus with standard of care therapies such as

chemotherapy or radiotherapy has a synergistic anti-cancer effect compared to using

the agents on their own [42].

Prostate cancer (PCa) is the most common cancer in men in the UK [358]. The first

line of standard care for patients with castration-resistant PCa (CRPC) is the

chemotherapeutic Docetaxel, which has shown to have a median survival benefit of 2

to 3 months [40]. A newer taxane analogue, Cabazitaxel, is sometimes provided as a

treatment for metastatic CRPC in patients who had previously been treated with

Docetaxel [182], as it displays activity in Docetaxel-resistant tumour cells [187, 188].

The 5-year survival rate is around 30% for patients whose disease was metastatic at

diagnosis [358]. Clearly more efficacious treatment strategies are required.

Taxane chemotherapy drugs display anti-cancer properties by binding to cellular

microtubules [182]. In turn, this inhibits microtubule disassembly and mitosis [183],

and ultimately leads to cell death. Reovirus also associates with and stabilises

microtubules for efficient viral growth [189-191], and PCa cell lines have been shown

to be susceptible to reovirus oncolysis [155]. The combination of reovirus and

Docetaxel promoted synergistic PCa cell death in vitro and in vivo, through increased

apoptosis, necrosis, microtubule stability and viral replication [192]. Conventional

chemotherapy involves treating patients with the maximum tolerated dose (MTD).

This is typically given in 3 week cycles, with extensive drug-free breaks in-between

to allow the patient to recover from the cytotoxic side-effects. However, these drug-

free periods allow tumour vasculature re-growth, leading to disease progression.

Therefore, the MTD method may not be the optimum way to administer such drugs.

Metronomic chemotherapy (MC) is defined as the frequent administration of

chemotherapy agents at doses below the MTD and with no prolonged drug-free

breaks. Besides targeting cancer cells, MC mainly kills endothelial cells involved in

201

tumour angiogenesis [197] such as circulating endothelial cells (CEC) in the tumour

vasculature and bone-marrow-derived endothelial progenitor cells (EPC) [195]. The

way in which MC enhances the anti-cancer immune response is mentioned in the

discussion of this chapter. MC may have a long-standing effect on preventing tumour

growth and has the potential benefit of being an inexpensive treatment [203].

Monotherapy with Paclitaxel or Docetaxel has demonstrated effective anti-cancer

responses in the metronomic setting [213, 214]. MC treatment with Cabazitaxel has

not yet been explored, nor has the combination of MC with reovirus or any other

oncolytic virus, which is the focus of this chapter.

202

6.2 STUDY OBJECTIVE

The objective of this chapter was to compare treatment of reovirus in combination

with either Cabazitaxel or Docetaxel in PCa cell lines. In vitro, we aimed to test these

combinations using the taxane drugs at the MTD (IC50) or at sub-lethal doses (<IC50)

to determine whether an anti-cancer effect could be achieved.


1. Determination of the reovirus, Cabazitaxel and Docetaxel IC50 values in a

selection of PCa cell lines, by the MTS assay and CalcuSyn software.

2. Assessment of the interaction between the concurrent combination of reovirus

and Cabazitaxel or Docetaxel at fixed-dose ratios in DU145 and LNCaP PCa cell

lines, by the MTS assay and the Chou and Talalay equation.

3. Comparison of concurrent and sequential combination treatments of reovirus and

Cabazitaxel in the DU145 cell line, by the MTS assay and the Chou and Talalay

equation.

4. Assessment of the interaction between the concurrent combination of reovirus

and Cabazitaxel or Docetaxel at non-fixed low doses in the DU145 cell line, by

the MTS assay and Bliss Independence analysis.

5. Measurement of acetylated α-tubulin in combination treated DU145 cell lysates

by western blotting, to determine whether microtubule stability is a factor that

contributes to a synergistic interaction at doses ≤IC50.

6. Measurement of intracellular and extracellular reovirus yield in DU145 cells by

the plaque assay, to determine whether viral replication or secretion is enhanced

following combination treatment at doses ≤IC50.

7. Assessment of DU145 cell survival by the MTS assay after using z-VAD-FMK

and Necrostatin-1, to determine the roles of apoptosis or necroptosis in the

synergistic effect caused by combination treatment.

203

6.3 RESULTS

6.3.1 Determination of reovirus, Cabazitaxel and Docetaxel IC50 values in PCa

cell lines

To assess the maximum tolerated dose (MTD) of reovirus, Cabazitaxel and Docetaxel

in PCa cell lines in vitro, the IC50 values were calculated. Three human PCa cell

lines (DU145, LNCaP and PC3), the TRAMP-C2 transgenic mouse PCa cell line, and

the human prostatic stromal myofibroblast WPMY-1 cell line (non-cancerous and

SV40 transformed), were treated with serial dilutions of reovirus, Cabazitaxel or

Docetaxel for 96 hours. Cells were also treated with media alone for the same time-

period as an untreated control. Cell survival was then analysed by the MTS assay

(Section 2.9).

Figure 6.1, 6.2 and 6.3 displays the % cell survival curves of each cell line treated

with reovirus, Cabazitaxel or Docetaxel respectively, relative to the untreated cells.

The IC50 values were determined by CalcuSyn software, which uses the median-effect

equation derived by Chou [217] (Section 2.27.4). Starting with the most sensitive cell

line, the order of susceptibility to reovirus was TRAMP-C2 (IC50 MOI 0.045), DU145

(IC50 MOI 2.850), LNCaP (IC50 MOI 5.410), PC3 (IC50 MOI 17.710) and WPMY-1

(IC50 MOI 47.300). The order of susceptibility to Cabazitaxel was DU145 (IC50

0.114µM), LNCaP (IC50 0.207µM), TRAMP-C2 (IC50 0.270µM), WPMY-1 (IC50

13.730µM) and PC3 (IC50 92.990µM). Comparatively, the order of susceptibility to

Docetaxel was DU145 (IC50 0.035µM), TRAMP-C2 (IC50 0.232µM), LNCaP (IC50

1.742µM), WPMY-1 (IC50 10.361µM) and PC3 (IC50 143.550µM).

204

Figure 6.1. Dose-response curves and IC50 values generated from cells infected with reovirus

alone. A. DU145, B. LNCaP, C. PC3, D. TRAMP-C2, and E. WPMY-1 were infected with serial

dilutions of reovirus for 96 hours before assessing the cell survival relative to the un-treated cells by the

MTS assay. Error bars represent SEM from three assay repeats. F. The reovirus IC50 values were

derived using CalcuSyn software. The table shows the mean of three assay repeats ± SEM. Starting

with the most sensitive cell line, the order of susceptibility to reovirus was TRAMP-C2, DU145,

LNCaP, PC3 and WPMY-1.

1.4

2.8

5.5

11.1

22.1

0

2 5

5 0

7 5

1 0 0

1 2 5

D U 1 4 5


% c

ell

su

rv

iva

l

1.4

2.7

5.4

10.8

21.6

0

2 5

5 0

7 5

1 0 0

1 2 5

L N C a P


% c

ell

su

rv

iva

l

1.6

3.1

6.3

12.5

25.0

0

2 5

5 0

7 5

1 0 0

1 2 5

P C 3


% c

ell

su

rv

iva

l

0.0

098

0.0

20

0.0

40

0.0

78

0.1

6

0

2 5

5 0

7 5

1 0 0

1 2 5

T R A M P -C 2


% c

ell

su

rv

iva

l

7.8

15.6

31.3

62.5

125.0

0

2 5

5 0

7 5

1 0 0

1 2 5

W P M Y -1


% c

ell

su

rv

iva

lA B

C D

E F

Cell line Reovirus IC50 (MOI) ± SEM

DU145 2.850 ± 0.420

LNCaP 5.410 ± 2.700

PC3 17.710 ± 3.630

TRAMP-C2 0.045 ± 0.004

WPMY-1 47.300 ± 3.210

205

Figure 6.2. Dose-response curves and IC50 values generated from cells infected with Cabazitaxel


dilutions of Cabazitaxel for 96 hours before assessing the cell survival relative to the un-treated cells by

the MTS assay. Error bars represent SEM from three assay repeats. F. The Cabazitaxel IC50 values

were derived using CalcuSyn software. The table shows the mean of three assay repeats ± SEM.

Starting with the most sensitive cell line, the order of susceptibility to Cabazitaxel was DU145, LNCaP,

TRAMP-C2, WPMY-1 and PC3.

0.0

28

0.0

56

0.1

1

0.2

2

0.4

4

0

2 5

5 0

7 5

1 0 0

1 2 5

D U 1 4 5

C a b a z ita x e l (µ M )

% c

ell

su

rv

iva

l

0.0

45

0.0

89

0.1

8

0.3

6

0.7

1

0

2 5

5 0

7 5

1 0 0

1 2 5

L N C a P


% c

ell

su

rv

iva

l

8.0

16.0

32.0

64.0

128.0

0

2 5

5 0

7 5

1 0 0

1 2 5

P C 3


% c

ell

su

rv

iva

l

0.0

94

0.1

9

0.3

8

0.7

5

1.5

0

0

2 5

5 0

7 5

1 0 0

1 2 5

T R A M P -C 2


% c

ell

su

rv

iva

l

3.1

6.3

12.5

25.0

50.0

0

2 5

5 0

7 5

1 0 0

1 2 5

W P M Y -1


% c

ell

su

rv

iva

lA B

C D

E F

Cell line Cabazitaxel IC50 (µM) ± SEM

DU145 0.114 ± 0.025

LNCaP 0.207 ± 0.029

PC3 92.990 ± 6.420

TRAMP-C2 0.270 ± 0.010

WPMY-1 13.730 ± 0.194

206

Figure 6.3. Dose-response curves and IC50 values generated from cells treated with Docetaxel


dilutions of Docetaxel for 96 hours before assessing the cell survival relative to the un-treated cells by

MTS assay. Error bars represent SEM from three assay repeats. F. The Docetaxel IC50 values were

derived using CalcuSyn software. The table shows the mean of three assay repeats ± SEM. Starting

with the most sensitive cell line, the order of susceptibility to Docetaxel was DU145, TRAMP-C2,

LNCaP, WPMY-1 and PC3.

0.0

10

0.0

21

0.0

41

0.0

83

0.1

7

0

2 5

5 0

7 5

1 0 0

1 2 5

D U 1 4 5

D o c e ta x e l (µ M )

% c

ell

su

rv

iva

l

0.2

50.5

1.0

2.0

4.0

0

2 5

5 0

7 5

1 0 0

1 2 5

L N C a P


% c

ell

su

rv

iva

l

40.0

80.0

160.0

320.0

640.0

0

2 5

5 0

7 5

1 0 0

1 2 5

P C 3


% c

ell

su

rv

iva

l

0.0

94

0.1

9

0.3

8

0.7

5

1.5

0

0

2 5

5 0

7 5

1 0 0

1 2 5

T R A M P -C 2


% c

ell

su

rv

iva

l

3.1

6.3

12.5

25.0

50.0

0

2 5

5 0

7 5

1 0 0

1 2 5

W P M Y -1


% c

ell

su

rv

iva

lA B

C D

E F

Cell line Docetaxel IC50 (µM) ± SEM

DU145 0.035 ± 0.009

LNCaP 1.742 ± 0.063

PC3 143.550 ± 8.390

TRAMP-C2 0.232 ± 0.009

WPMY-1 10.361 ± 0.087

207

6.3.2 Concurrent combination of reovirus with Docetaxel or Cabazitaxel mostly

had an anti-cancer synergistic effect in PCa cell lines, as determined by

the Chou and Talalay equation

Next, the effect of concurrently combining reovirus with Cabazitaxel or Docetaxel

was assessed (Section 2.23.1). DU145 and LNCaP PCa cell lines were treated with

reovirus, Cabazitaxel or Docetaxel alone, at doses representing 0.25, 0.5, 1, 2 and 4

fold the IC50 values. Cells were also treated with reovirus in combination with

Cabazitaxel, or reovirus in combination with Docetaxel, at doses representing 0.25,

0.5, 1, 2 and 4 fold the IC50 values. As an untreated control, cells were incubated with

media alone. After 96 hours, the cell survival was assessed by the MTS assay

(Section 2.9) and the combination index (CI) values were determined using CalcuSyn

software, which measures the degree of interaction between two agents by the CI

equation of Chou and Talalay [234] (Section 2.27.5). A CI value of 1 denoted an

additive interaction, <1 a synergistic interaction, and >1 an antagonistic interaction.

Figure 6.4 A, 6.5 A, 6.6 A and 6.7 A show the % cell survival curves of DU145 and

LNCaP PCa cell lines treated with reovirus, Cabazitaxel or Docetaxel as single

agents, or with reovirus in combination with either taxane drug, relative to the

untreated cells. For both cell lines, both combinations displayed a synergistic effect at

the effective dose 50 (ED50) and ED75 (Figure 6.4 C, 6.5, 6.6 C and 6.7 C). This is

depicted on the isobologram curves where the combination data points are located

below the lines of additivity (Figure 6.4 B, 6.5 B, 6.6 B and 6.7 B). This suggested

that the combinations had more effect on cell death than each agent used

independently of the other. When comparing the synergistic effect between the two

combination treatments in DU145 cells, reovirus and Docetaxel was marginally more

synergistic than reovirus and Cabazitaxel at ED50, ED75 and ED90 (CI=0.44, 0.33, 0.30

and CI=0.61, 0.38, 0.31, respectively). In the LNCaP cell line, reovirus and

Docetaxel showed strong synergism at the ED50 (CI=0.22), whereas reovirus and

Cabazitaxel showed a slightly lower synergistic effect (CI=0.44). However, at ED75,

both combinations showed a similar level of synergy (CI=0.39 and CI=0.41). At

ED90, reovirus and Docetaxel displayed moderate antagonism (CI=1.23), as depicted

on the isobologram where the data point is located above the line of additivity, whilst

reovirus and Cabazitaxel showed synergism (CI=0.37).

208

A

B

C

Figure 6.4. Concurrent combination of reovirus and Cabazitaxel at doses 4, 2, 1, 0.5 and 0.25

fold the IC50 had a synergistic anti-cancer effect in the DU145 PCa cell line. A. % Cell survival

was assessed by the MTS assay after 96 hours incubation with reovirus alone (orange circles),

Cabazitaxel alone (blue squares) or reovirus and Cabazitaxel in combination (purple triangles), relative

to the un-treated cells. Error bars represent SEM from three assay repeats. B. The isobologram curve

was generated using CalcuSyn software and shows the combination data points located below the line

of additivity of the effective dose 50 (ED50), ED75 and ED90, which indicated synergism. C. The

combination index (CI) values of the ED50, ED75 and ED90 displayed a synergistic anti-cancer effect, as

determined using CalcuSyn software. The CI ± SEM of three assay repeats is shown.

0.2

5×IC

50

0.5

×IC

50

1×IC

50

2×IC

50

4×IC

50

0

2 5

5 0

7 5

1 0 0

1 2 5

R e o v iru s (M O I) , C a b a z ita x e l (µ M )

% c

ell

su

rv

iva

l

R e o

C a b

R e o + C a b

DU145: Reo + Cab

ED CI ± SEM Combination effect

ED50 0.61 ± 0.071 +++ Synergism

ED75 0.38 ± 0.053 +++ Synergism

ED90 0.31 ± 0.024 +++ Synergism

209

A

B

C

Figure 6.5. Concurrent combination of reovirus and Docetaxel at doses 4, 2, 1, 0.5 and 0.25 fold

the IC50 had a synergistic anti-cancer effect in the DU145 PCa cell line. A. % Cell survival was

assessed by the MTS assay after 96 hours incubation with reovirus alone (orange circles), Docetaxel

alone (blue squares) or reovirus and Docetaxel in combination (purple triangles), relative to the un-

treated cells. Error bars represent SEM from three assay repeats. B. The isobologram curve was

generated using CalcuSyn software and shows the combination data points located below the line of

additivity of the effective dose 50 (ED50), ED75 and ED90, which indicated synergism. C. The



DU145: Reo + Doc


ED50 0.44 ± 0.055 +++ Synergism

ED75 0.33 ± 0.030 +++ Synergism

ED90 0.30 ± 0.019 +++ Synergism

0.2

5×IC

50

0.5

×IC

50

1×IC

50

2×IC

50

4×IC

50

0

2 5

5 0

7 5

1 0 0

1 2 5

R e o v iru s (M O I) / D o c e ta x e l (µ M )

% c

ell

su

rv

iva

l

R e o

D o c

R e o + D o c

210

A

B

C

Figure 6.6. Concurrent combination of reovirus and Cabazitaxel at doses 4, 2, 1, 0.5 and 0.25

fold the IC50 had a synergistic anti-cancer effect in the LNCaP PCa cell line. A. % Cell survival

was assessed by the MTS assay after 96 hours incubation with reovirus alone (orange circles),

Cabazitaxel alone (blue squares) or reovirus and Cabazitaxel in combination (purple triangles), relative

to the un-treated cells. Error bars represent SEM from three assay repeats. B. The isobologram curve

was generated using CalcuSyn software and shows the combination data points located below the line

of additivity of the effective dose 50 (ED50), ED75 and ED90, which indicated synergism. C. The



LNCaP: Reo + Cab


ED50 0.44 ± 0.031 +++ Synergism

ED75 0.41 ± 0.012 +++ Synergism

ED90 0.37 ± 0.009 +++ Synergism

0.2

5×IC

50

0.5

×IC

50

1×IC

50

2×IC

50

4×IC

50

0

2 5

5 0

7 5

1 0 0

1 2 5


% c

ell

su

rv

iva

l

R e o

C a b

R e o + C a b

211

A

B

C

Figure 6.7. Concurrent combination of reovirus and Docetaxel at doses 4, 2, 1, 0.5 and 0.25 fold

the IC50 had a synergistic anti-cancer effect in the LNCaP PCa cell line at the ED50 and ED75, but

not at ED90. A. % Cell survival was assessed by the MTS assay after 96 hours incubation with

reovirus alone (orange circles), Docetaxel alone (blue squares) or reovirus and Docetaxel in

combination (purple triangles), relative to the un-treated cells. Error bars represent SEM from three

assay repeats. B. The isobologram curve was generated using CalcuSyn software and shows the

combination data points for ED50 and ED75 below the line of additivity, which indicated synergism.

However, the combination data point for ED90 exhibited antagonism, as it was located above the line of

additivity. C. The combination index (CI) values of the ED50 and ED75 displayed a synergistic anti-

cancer effect, whereas the CI value for ED90 showed moderate antagonism, as determined using

CalcuSyn software. The CI ± SEM of three assay repeats is shown.

0.2

5×IC

50

0.5

×IC

50

1×IC

50

2×IC

50

4×IC

50

0

2 5

5 0

7 5

1 0 0

1 2 5

R e o v iru s (M O I) , D o c e ta x e l (µ M )

% c

ell

su

rv

iva

l

R e o

D o c

R e o + D o c

LNCaP: Reo + Doc


ED50 0.22 ± 0.091 ++++ Strong Synergism

ED75 0.39 ± 0.042 +++ Synergism

ED90 1.23 ± 0.150 - - Moderate Antagonism

212

6.3.3 Concurrent combination of reovirus and Cabazitaxel resulted in greater

synergistic anti-cancer activity in the DU145 PCa cell line than sequential

combination treatment

In order to assess the most synergistic sequencing strategy (Section 2.23.2), we

concurrently treated DU145 cells with reovirus and Cabazitaxel at doses representing

1.00, 0.50, 0.25, 0.13 and 0.06 ×IC50, or treated sequentially at these doses with

Cabazitaxel alone for 1 hour (sequential treatment 1), 4 hours (sequential treatment 2)

or 24 hours (sequential treatment 3) before adding reovirus for a total of 96 hours.

Alternatively, cells were treated at each dose with reovirus alone for 24 hours

(sequential treatment 4) or 48 hours (sequential treatment 5) before adding

Cabazitaxel for the remaining time period totalling 96 hours. Cells were also treated

with each dose of reovirus, Cabaziatxel or Docetaxel as single agents for the indicated

time periods. The cell survival was then assessed by the MTS assay (Section 2.9) and

the combination index (CI) values were determined using CalcuSyn software via the

Chou and Talalay equation [234] (Section 2.27.5).

Figure 6.8 A shows the % cell survival curves of DU145 cells treated with the

concurrent combination of reovirus and Cabazitaxel, and the five different sequential

combination treatments, relative to untreated cells. Concurrent combination treatment

had a greater synergistic interaction than sequential combinations 1, 2, 3, 4 and 5 at

the ED50, ED75 and ED90 (Figure 6.8 B). Therefore, it was decided that all future

combination treatments would be performed concurrently.

213

A

0.0

6×IC

50

0.1

3×IC

50

0.2

5×IC

50

0.5

0×IC

50

1.0

0×IC

50

0

2 5

5 0

7 5

1 0 0

1 2 5


% c

ell

su

rv

iva

l

S e q u e n tia l tre a tm e n t 1





C o n c u rre n t tre a tm e n t

214

B

Figure 6.8. Concurrent combination treatment of reovirus and Cabazitaxel resulted in a more

efficient anti-cancer synergistic interaction than sequential combination treatment, in the DU145

PCa cell line. A. Cells were treated concurrently with reovirus and Cabazitaxel for 96 hours (black

open circles), or treated sequentially with Cabazitaxel for 1 hour (sequential treatment 1, green circles),

4 hours (sequential treatment 2, blue squares) or 24 hours (sequential treatment 3, red triangles) before

adding reovirus for a total of 96 hours. Cells were also treated with reovirus for 24 hours (sequential

treatment 4, orange triangles) or 48 hours (sequential treatment 5, purple diamonds), before adding

Cabazitaxel for a total of 96 hours. The cell survival in each treatment was assessed by the MTS assay

relative to the untreated cells. Error bars represent SD from triplicate samples. B. The combination

index (CI) values of the ED50, ED75 and ED90 were determined using CalcuSyn software.

Concurrent treatment Sequential treatment 1

ED CI Combination effect CI Combination effect

ED50 0.23 ++++

Strong synergism 0.56

+++ Synergism

ED75 0.09 +++++

Very strong synergism 0.25

++++ Strong synergism

ED90 0.04 +++++





ED50 0.27 ++++


++ Moderate synergism

ED75 0.23 ++++


- Slight antagonism

ED90 0.28 ++++


--- Antagonism



ED50 0.07 +++++


+++ Synergism

ED75 0.05 +++++



ED90 0.03 +++++





ED50 0.29 ++++



ED75 0.28 ++++


+++ Synergism

ED90 0.29 ++++


+++ Synergism



ED50 0.15 ++++


+++ Synergism

ED75 0.10 ++++



ED90 0.06 +++++


+++++ Very strong synergism

215

6.3.4 Combination of reovirus with Cabazitaxel or Docetaxel at doses below the

IC50 values had an anti-cancer synergistic effect on the DU145 PCa cell

line, as determined by the Bliss Independence model

We then assessed the effect of combining reovirus with Cabazitaxel at a greater range

of doses, including doses much lower than the IC50 value of each agent (Section

2.23.3). The DU145 PCa cell line was treated with reovirus or Cabazitaxel alone, or

in combination for 96 hours. Each dose of Cabazitaxel representing 0.05, 0.11, 0.22,

0.44, 0.88, 1.75, 2.63 and 3.51 ×IC50 (IC50 0.114µM), was combined with various

doses of reovirus representing 0.13, 0.26, 0.44, 0.88, 1.75, 3.51 and 7.02 ×IC50 (IC50

MOI 2.85). Thus, in this experiment the combinations were not set at fixed ratios.

Cells were also treated with media alone as an untreated control. The cell survival

was evaluated by the MTS assay (Section 2.9) and the interaction between the two

agents was then assessed by the Bliss Independence analysis [224] (Section 2.27.6).

The Bliss analysis spreadsheet was provided to us by Professor Kevin Harrington’s

laboratory, with permission from MedImmune LLC, USA [223].

Figure 6.9 A shows the % cell survival curves of DU145 cells treated with reovirus or

Cabazitaxel as single agents, or with reovirus in combination with Cabazitaxel,

relative to the untreated cells. Figure 6.9 B and C display a table and contour map of

the ΔE values for each combination, as well as the Bliss analysis range (the upper and

lower confidence intervals (Ci)). A positive ΔE value indicated synergism, whereas a

negative ΔE value indicated antagonism. When ΔE=0, the combination was

considered to have Bliss Independence (or addition). The contour map and table are

colour coded according to the level of interaction between reovirus and Cabazitaxel

and all values are expressed as a percentage (i.e. a 20% synergistic effect equates to

ΔE=0.20). Values highlighted in orange show the greatest synergistic effect (40% to

50%), whereas values highlighted in black show an additive or antagonistic

interaction (-10% to 0%). All combinations demonstrated a synergistic effect apart

from the lowest doses of reovirus and Cabazitaxel (0.13× and 0.05× IC50 of reovirus

and Cabazitaxel respectively). Thus, the data suggested that concentrations of

reovirus 0.26× lower than the IC50, and concentrations of Cabazitaxel 0.11× lower

than the IC50, were capable of generating a synergistic anti-cancer effect compared to

treating DU145 cells with reovirus or Cabazitaxel as single agents.

216

The effect of combining reovirus with Docetaxel at a wide range of non-fixed dose

ratios was then determined (Section 2.23.3). DU145 cells were treated with reovirus

or Docetaxel alone, or in combination for 96 hours. Each dose of Docetaxel

representing 0.07, 0.14, 0.29, 0.57, 0.86, 2.86 and 5.71 ×IC50 (IC50 0.035µM), was

combined with various doses of reovirus representing 0.13, 0.26, 0.44, 0.88, 1.75,

3.51 and 7.02 ×IC50 (IC50 MOI 2.85). Additionally, cells were treated with media

alone as an untreated control. The cell survival was evaluated by the MTS assay

(Section 2.9) and the interaction between the two agents was then assessed by the

Bliss Independence analysis (Section 2.27.6).

Figure 6.10 A shows the % cell survival curves of DU145 cells treated with reovirus

or Docetaxel as single agents, or with reovirus in combination with Docetaxel,

relative to the untreated cells. The contour map of the ΔE values for each

combination are shown in Figure 6.10 C. The ΔE values and the Bliss analysis range

are displayed in Figure 6.10 B. All combinations showed a synergistic effect. This

implied that concentrations of reovirus 0.13× lower than the IC50, and concentrations

of Docetaxel 0.07× lower than the IC50, were capable of generating a synergistic anti-

cancer effect compared to treating DU145 cells with reovirus or Docetaxel as single

agents.

Therefore, when comparing the two combination treatments using the Bliss

Independence model in the DU145 PCa cell line, it appears that reovirus and

Docetaxel can be combined at slightly lower doses than reovirus and Cabazitaxel,

whist still maintaining a level of anti-cancer synergism. However, there is a region

where reovirus and Cabazitaxel had a higher level of synergy of 46.56% (0.44×IC50

reovirus and 0.22×IC50 Cabazitaxel) compared to the combination of reovirus and

Docetaxel, which reached a maximum synergistic effect of 29.00% (1.75×IC50

reovirus and 0.07×IC50 Docetaxel). Overall, the data indicates that both combinations

have the potential to be used as metronomic treatments for PCa.

217

A

0.0

0 ×

IC

50

0.0

7 ×

IC

50

0.1

4 ×

IC

50

0.2

9 ×

IC

50

0.5

7 ×

IC

50

0.8

6 ×

IC

50

2.8

6 ×

IC

50

5.7

1 ×

IC

50

0

2 5

5 0

7 5

1 0 0

1 2 5


% c

ell

su

rv

iva

l

0 .0 0 × IC 5 0

0 .0 5 × IC 5 0

0 .1 1 × IC 5 0

0 .2 2 × IC 5 0

0 .4 4 × IC 5 0

0 .8 8 × IC 5 0

1 .7 5 × IC 5 0

2 .6 3 × IC 5 0

3 .5 1 × IC 5 0

Ca

ba

zit

ax

el

(µM

)

218

B C

Figure 6.9. Combination treatment of the DU145 PCa cell line with reovirus and Cabazitaxel at doses much lower than the IC50 values mostly had an anti-

cancer synergistic effect, as determined by Bliss Independence analysis. Cells were treated with Cabazitaxel at doses representing 0.05, 0.11, 0.22, 0.44, 0.88,

1.75, 2.63 and 3.51 fold the IC50 (IC50 0.114µM), combined with reovirus at doses representing 0.13, 0.26, 0.44, 0.88, 1.75, 3.51 and 7.02 fold the IC50 (IC50 MOI

2.85). Alternatively, cells were treated with Cabazitaxel or reovirus as single agents at the stated doses, or with media alone as an untreated control. A. After 96

hours, the % cell survival in each treatment was assessed by the MTS assay, relative to the untreated cells. The survival curves display error bars that represent SEM

from two assay repeats. B. The interaction between Cabazitaxel and reovirus was assessed by the Bliss Independence analysis spreadsheet (MedImmune LLC, USA

[223]. The ΔE values for each combination and the Bliss analysis range (the upper and lower confidence intervals (Ci)) are displayed in the table, which is colour

coded according to the level of interaction. Orange shows the highest level of synergism whereas black shows an additive or antagonistic interaction. All

combinations demonstrated a synergistic anti-cancer effect of up to 46.56%, apart from the lowest dose of reovirus and Cabazitaxel. C. The ΔE values for each

combination of Cabazitaxel and reovirus were plotted onto a contour map, enabling easy visualisation of the level of interaction.

7.02 × IC50 3.51 × IC50 1.75 × IC50 0.88 × IC50 0.44 × IC50 0.26 × IC50 0.13 × IC50

ΔE 5.83% 6.06% 7.17% 10.15% 12.59% 10.59% 5.36%

Upper Ci 7.21% 6.91% 9.09% 11.59% 15.76% 13.77% 8.88%

Lower Ci 4.55% 5.22% 5.36% 8.76% 9.54% 7.48% 1.97%

ΔE 5.51% 5.84% 7.06% 9.58% 11.99% 12.02% 6.79%

Upper Ci 6.96% 6.97% 8.84% 10.89% 15.47% 14.64% 10.34%

Lower Ci 4.12% 4.72% 5.35% 8.32% 8.61% 9.44% 3.34%

ΔE 3.70% 4.11% 5.66% 7.45% 8.79% 6.78% 4.33%

Upper Ci 4.53% 4.64% 6.79% 8.28% 10.77% 8.72% 7.56%

Lower Ci 2.88% 3.58% 4.54% 6.62% 6.82% 4.85% 1.11%

ΔE 8.98% 8.67% 12.16% 17.51% 21.25% 21.72% 14.31%

Upper Ci 11.34% 10.02% 15.55% 20.11% 26.10% 25.45% 19.14%

Lower Ci 6.76% 7.36% 8.95% 14.99% 16.61% 18.08% 9.70%

ΔE 13.77% 14.46% 19.39% 26.65% 33.90% 36.87% 27.48%

Upper Ci 16.25% 15.97% 23.02% 29.50% 38.66% 40.96% 34.05%

Lower Ci 11.44% 12.98% 15.94% 23.88% 29.36% 32.88% 21.13%

ΔE 20.14% 20.39% 26.84% 36.05% 46.56% 42.97% 25.63%

Upper Ci 23.25% 23.42% 32.00% 41.92% 56.19% 52.03% 34.95%

Lower Ci 17.15% 17.38% 21.84% 30.25% 37.10% 34.00% 16.50%

ΔE 22.53% 20.72% 20.82% 25.78% 36.11% 27.04% 8.25%

Upper Ci 28.26% 28.35% 31.34% 34.08% 44.61% 37.62% 19.01%

Lower Ci 17.04% 13.13% 10.58% 17.58% 27.93% 16.61% -2.19%

ΔE 14.74% 9.86% 13.17% 8.30% 12.17% 6.78% -3.30%

Upper Ci 20.05% 11.82% 19.42% 12.23% 20.44% 10.01% 3.65%

Lower Ci 9.49% 7.91% 7.00% 4.40% 3.98% 3.58% -10.16%

Reovirus (MOI)

Cab

azit

axel

(µ

M)

3.51 × IC50

2.63 × IC50

1.75 × IC50

0.88 × IC50

0.44 × IC50

0.22 × IC50

0.11 × IC50

0.05 × IC50

219

A

0.0

0 ×

IC

50

0.1

3 ×

IC

50

0.2

6 ×

IC

50

0.4

4 ×

IC

50

0.8

8 ×

IC

50

1.7

5 ×

IC

50

3.5

1 ×

IC

50

7.0

2 ×

IC

50

0

2 5

5 0

7 5

1 0 0

1 2 5


% c

ell

su

rv

iva

l

0 .0 0 × IC 5 0

0 .0 7 × IC 5 0

0 .1 4 × IC 5 0

0 .2 9 × IC 5 0

0 .5 7 × IC 5 0

0 .8 6 × IC 5 0

2 .8 6 × IC 5 0

5 .7 1 × IC 5 0

Do

ce

tax

el

(µM

)

220

B C

Figure 6.10. Combination treatment of the DU145 PCa cell line with reovirus and Docetaxel at doses much lower than the IC50 values had an anti-cancer

synergistic effect, as determined by Bliss Independence analysis. Cells were treated with Docetaxel at doses representing 0.07, 0.14, 0.29, 0.57, 0.86, 2.86 and 5.71 fold

the IC50 (IC50 0.035µM), combined with reovirus at doses representing 0.13, 0.26, 0.44, 0.88, 1.75, 3.51 and 7.02 fold the IC50 (IC50 MOI 2.85). Alternatively, cells were

treated with Docetaxel or reovirus as single agents at the stated doses, or with media alone as an untreated control. A. After 96 hours the % cell survival in each treatment

was assessed by the MTS assay, relative to the untreated cells. The survival curves display error bars that represent SEM from two assay repeats. B. The interaction

between Docetaxel and reovirus was assessed by the Bliss Independence analysis spreadsheet (MedImmune LLC, USA [223]. The ΔE values for each combination and the

Bliss analysis range (the upper and lower confidence intervals (Ci)) are displayed in the table, which is colour coded according to the level of interaction, where purple shows

the highest level of synergism. All combinations demonstrated a synergistic anti-cancer effect of up to 29.00%. C. The ΔE values for each combination of Docetaxel and

reovirus were plotted onto a contour map, enabling easy visualisation of the level of interaction.

7.02 × IC50 3.51 × IC50 1.75 × IC50 0.88 × IC50 0.44 × IC50 0.26 × IC50 0.13 × IC50

ΔE 2.66% 2.48% 5.56% 6.52% 9.00% 9.05% 8.31%

Upper Ci 3.81% 3.83% 8.66% 10.13% 12.26% 11.84% 12.31%

Lower Ci 1.53% 1.18% 2.64% 3.14% 5.86% 6.41% 4.46%

ΔE 2.20% 2.97% 5.36% 5.52% 6.94% 7.48% 5.53%

Upper Ci 3.33% 4.15% 7.64% 8.67% 9.55% 10.74% 8.92%

Lower Ci 1.10% 1.83% 3.24% 2.61% 4.43% 4.36% 2.30%

ΔE 4.53% 4.87% 10.22% 10.59% 13.10% 11.95% 10.27%

Upper Ci 7.01% 6.91% 15.03% 17.14% 19.87% 18.55% 21.30%

Lower Ci 2.14% 2.96% 5.94% 4.75% 6.68% 5.78% -0.27%

ΔE 5.88% 6.30% 11.16% 11.50% 13.76% 15.62% 11.10%

Upper Ci 7.42% 8.61% 17.29% 18.75% 20.02% 22.97% 21.20%

Lower Ci 4.44% 4.13% 5.60% 5.00% 7.88% 8.72% 1.52%

ΔE 8.59% 10.33% 19.85% 20.52% 22.27% 25.27% 17.80%

Upper Ci 11.12% 13.10% 26.46% 28.89% 27.99% 35.42% 25.97%

Lower Ci 6.12% 7.65% 13.60% 12.63% 16.79% 15.41% 9.95%

ΔE 9.78% 11.85% 21.45% 14.96% 19.03% 19.94% 8.58%

Upper Ci 12.84% 15.42% 30.85% 25.35% 29.49% 31.45% 18.12%

Lower Ci 6.76% 8.36% 12.34% 4.96% 8.76% 8.67% -0.69%

ΔE 11.80% 16.72% 29.00% 14.44% 22.34% 19.29% 3.65%

Upper Ci 17.14% 22.59% 44.14% 28.72% 37.52% 36.97% 22.67%

Lower Ci 6.63% 11.11% 14.86% 1.47% 7.81% 2.40% -14.47%

Reovirus (MOI)

Do

ceta

xel (

µM

)

5.71 × IC50

2.86 × IC50

0.86 × IC50

0.57 × IC50

0.29 × IC50

0.14 × IC50

0.07 × IC50

221

6.3.5 The synergistic anti-cancer effect of the combination of reovirus and

Cabazitaxel or Docetaxel at the IC50 doses may be due to enhanced

microtubule stability

We next intended to determine the mechanism of the synergistic anti-cancer

interaction of the combination of reovirus and Cabazitaxel or Docetaxel in the DU145

PCa cell line. The combination of Docetaxel and reovirus was previously shown to

enhance microtubule stability and correlated with promotion of cell death [192].

Therefore, cell lysates were collected from DU145 cells after 24 hours treatment with

reovirus, Cabazitaxel or Docetaxel alone at the IC50 doses, or with reovirus in

combination with either taxane drug at the IC50 doses. Lysates were also prepared

from cells treated with media alone. The level of acetylated α-tubulin in each sample

was quantified by western blotting and densitometry analysis (Section 2.14), which is

proportional to, and is used as a marker of microtubule stability.

As expected, all treatments caused an increase in acetylated α-tubulin compared to

cells treated with media alone. Cabazitaxel appeared to be a more efficient

microtubule stabiliser than Docetaxel when used as single agent treatments. The

greatest increase in acetylated α-tubulin was observed in the combinations, with

reovirus and Cabazitaxel showing a slightly more intense band than reovirus and

Docetaxel (Figure 6.11 A and B). This suggested that the synergistic anti-cancer

effect associated with combination treatment is partly due to increased microtubule

stability.

222

A

B

Figure 6.11. Microtubule stability was enhanced after combination treatment with reovirus and

Cabazitaxel or Docetaxel at the IC50 doses, in the DU145 PCa cell line. Lysates were collected

from cells treated with reovirus, Docetaxel or Cabazitaxel as single agents, or with reovirus in

combination with Docetaxel or Cabazitaxel for 24 hours. Lysates were also prepared from cells treated

with media alone. A. The level of acetylated α-tubulin in each sample was determined by western

blotting. B. The intensity of the acetylated α-tubulin bands were quantified and normalised to their

corresponding β-actin bands by densitometry analysis. All treatments caused an increase in acetylated

α-tubulin compared to untreated (media alone) cells. Reovirus and Cabazitaxel in combination resulted

in a slightly greater amount of acetylated α-tubulin than the combination of reovirus and Docetaxel.

Reo

(1×IC

50 )

Do

c (

1×IC

50 )

Cab

(1×IC

50 )

Reo

(1×IC

50 )

+ D

oc (

1×IC

50 )

Reo

(1×IC

50 )

+ C

ab

(1×IC

50 )

Med

ia

0

2 0

4 0

6 0

8 0

1 0 0

Re

lati

ve

de

ns

ity

223

6.3.6 The synergistic anti-cancer effect of reovirus in combination with low

doses of Cabazitaxel or Docetaxel may be due to increased microtubule

stabilisation

Having shown that the combination of reovirus and Cabazitaxel or Docetaxel at low

doses had an anti-cancer synergistic effect, we determined whether enhanced

microtubule stability contributed to this. Firstly, lysates were collected from DU145

cells treated with reovirus at the IC50 dose (IC50 MOI 2.85) in combination with serial

dilutions of either taxane drug (1.0, 0.50, 0.25, 0.13, 0.06 and 0.03 ×IC50), for 24

hours. Lysates were also prepared from cells treated with media alone. The level of

acetylated α-tubulin in each sample was quantified by western blotting and

densitometry analysis (Section 2.14).

The western blot images in Figure 6.12 A and Figure 6.13 A show that all

combination treatments resulted in a greater intensity acetylated α-tubulin band than

cells treated with Cabazitaxel or Docetaxel alone at the IC50 doses. This was

confirmed by densitometry analysis (Figure 6.12 B and 6.13 B). Therefore, reovirus

(at the IC50 dose) in combination with Cabazitaxel or Docetaxel at doses as low as

0.03 ×IC50, caused enhanced microtubule stabilisation compared to single agent

treatment. This may partially explain the synergistic anti-cancer interaction observed

following combination treatment of reovirus and low doses of taxane drug.

224

A

B


Cabazitaxel at low doses, in the DU145 PCa cell line. Lysates were collected from cells treated with

media alone, Cabazitaxel alone, or with reovirus (at the IC50 dose) in combination with Cabazitaxel at

1.0, 0.50, 0.25, 0.13, 0.06 or 0.03 ×IC50 dose, for 24 hours. A. The level of acetylated α-tubulin in each

sample was determined by western blotting. B. The intensity of the acetylated α-tubulin bands were

quantified and normalised to their corresponding β-actin bands by densitometry analysis. All

combination treatments caused a greater increase in acetylated α-tubulin than cells treated with

Cabazitaxel as a single agent or with media alone.

Reo

(1.0

×IC

50 )

+ C

ab

(1.0

×IC

50 )

Reo

(1.0

×IC

50 )

+ C

ab

(0.5

0 ×

IC50 )

Reo

(1.0

×IC

50 )

+ C

ab

(0.2

5 ×

IC50 )

Reo

(1.0

×IC

50 )

+ C

ab

(0.1

3 ×

IC50 )

Reo

(1.0

×IC

50 )

+ C

ab

(0.0

6 ×

IC50 )

Reo

(1.0

×IC

50 )

+ C

ab

(0.0

3 ×

IC50 )

Cab

(1.0

0 ×

IC50 )

med

ia a

lon

e

0

1 0 0

2 0 0

3 0 0

4 0 0

Re

lati

ve

de

ns

ity

225

A

B


Docetaxel at low doses, in the DU145 PCa cell line. Lysates were collected from cells treated with

media alone, Docetaxel alone, or with reovirus (at the IC50 dose) in combination with Docetaxel at 1.0,

0.50, 0.25, 0.13, 0.06 or 0.03 ×IC50 dose, for 24 hours. A. The level of acetylated α-tubulin in each

sample was determined by western blotting. B. The intensity of the acetylated α-tubulin bands were

quantified and normalised to their corresponding β-actin bands by densitometry analysis. All

combination treatments caused a greater increase in acetylated α-tubulin than cells treated with

Docetaxel as a single agent or with media alone.

Reo

(1.0

×IC

50 )

+ D

oc (

1.0

×IC

50 )

Reo

(1.0

×IC

50 )

+ D

oc (

0.5

0 ×

IC50 )

Reo

(1.0

×IC

50 )

+ D

oc (

0.2

5 ×

IC50 )

Reo

(1.0

×IC

50 )

+ D

oc (

0.1

3 ×

IC50 )

Reo

(1.0

×IC

50 )

+ D

oc (

0.0

6 ×

IC50 )

Reo

(1.0

×IC

50 )

+ D

oc (

0.0

3 ×

IC50 )

Do

c (

1.0

0 ×

IC50 )

med

ia a

lon

e

0

2 5

5 0

7 5

1 0 0

1 2 5

Re

lati

ve

de

ns

ity

226

6.3.7 The synergistic anti-cancer effect of reovirus and Cabazitaxel or Docetaxel

in combination was not due to enhanced viral replication

We then determined whether increased viral replication contributes to the anti-cancer

synergistic effect caused by the combination of reovirus and Cabazitaxel or

Docetaxel. The reason for testing this was due to the fact that some studies have

reported enhanced viral titres in cancer cells treated with the co-administration of

reovirus and chemotherapeutic drugs [141, 192], whilst other studies have not found

this connection [335, 352, 359]. We treated the DU145 PCa cell line with reovirus

alone at the IC50 dose, or with reovirus (at the IC50) in combination with Cabazitaxel

or Docetaxel at concentrations representing 1.00, 0.25 and 0.06 ×IC50. Intracellular

and extracellular virus samples were collected after 24, 48 and 72 hours treatment and

the virus titre was determined by one-step growth curve analysis via the plaque assay

(Section 2.24).

Treatment with reovirus alone produced higher amounts of intracellular virus

compared to all combination treatments with Cabazitaxel (Figure 6.14 A). By 72

hours, the viral yield was approximately 1-log higher in cells treated with reovirus

alone than the combination of reovirus and Cabazitaxel at 1.00 ×IC50 dose, which

suggested that viral replication did not contribute to synergistic cancer cell kill.

Despite being lower in titre than reovirus alone, it was interesting to find that the

higher the concentration of Cabazitaxel used in combination with reovirus, the lower

the viral titre observed after 24 hours (although the treatment means were not

statistically different from each other by one-way ANOVA and Tukey’s post-hoc

test). This implied that the mode of cell death may be compromised according to the

dose of Cabazitaxel used when combined with reovirus. A similar trend was observed

when DU145 cells were treated with reovirus and Docetaxel at the same combination

ratios (Figure 6.15 A), although the intracellular viral titre was not as well spread in

comparison to treatment with reovirus and Cabazitaxel. There was no clear separation

in the extracellular viral growth curves of cells treated with reovirus alone and

combination treated cell supernatants, indicating that there was no significant

difference in viral release between treatment groups (Figure 6.14 B and 6.15 B).

227

A

B

Figure 6.14. Combination treatment of the DU145 PCa cell line with reovirus and Cabazitaxel

did not enhance the intracellular or extracellular viral yield compared to single agent reovirus

treatment. Cells were treated with reovirus alone at 1.00 ×IC50 dose (orange circles), or with reovirus

at 1.00 ×IC50 in combination with Cabazitaxel at 1.00 ×IC50 (purple squares), 0.25 ×IC50 (green

triangles) or 0.06 ×IC50 (blue triangles). A. Intracellular and B. Extracellular virus samples were

collected after 24, 48 and 72 hours. The virus titre was determined by one-step growth curve analysis

via the plaque assay. Reovirus alone produced higher amounts of intracellular virus compared to

combination treatment with Cabazitaxel, which was most evident at 72 hours when 1.00 ×IC50

Cabazitaxel was used, implying that viral replication was not a major factor that contributed to synergy.

The overlap in the extracellular viral growth curves suggested a lack of a relationship between

enhanced anti-cancer synergism and secretion of reovirus in the combination treatments. Viral titre is

shown on a log10 scale. The graphs show the mean of two assay repeats and error bars represent SEM.

024

48

72

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

1 0 1 0

1 0 1 1

1 0 1 2

In tra c e llu la r

T im e (h o u rs )

pfu

/mL

R e o (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + C a b (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + C a b (0 .2 5 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + C a b (0 .0 6 × IC 5 0 )

024

48

72

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

1 0 1 0

1 0 1 1

1 0 1 2

E x tra c e llu la r

T im e (h o u rs )

pfu

/mL

R e o (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + C a b (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + C a b (0 .2 5 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + C a b (0 .0 6 × IC 5 0 )

228

A

B

Figure 6.15. Combination treatment of the DU145 PCa cell line with reovirus and Docetaxel did

not enhance the intracellular or extracellular viral yield compared to single agent reovirus

treatment. Cells were treated with reovirus alone at 1.00 ×IC50 dose (orange circles), or with reovirus

at 1.00 ×IC50 in combination with Docetaxel at 1.00 ×IC50 (purple squares), 0.25 ×IC50 (green triangles)

or 0.06 ×IC50 (blue triangles). A. Intracellular and B. Extracellular virus samples were collected after

24, 48 and 72 hours. The virus titre was determined by one-step growth curve analysis via the plaque

assay. Combination treatment did not enhance the intracellular reovirus yield compared to when

reovirus was used as a single agent, implying that viral replication was not a major factor that

contributed to synergy. The overlap in the extracellular viral growth curves suggested a lack of a

relationship between enhanced anti-cancer synergism and secretion of reovirus in the combination

treatments. Viral titre is shown on a log10 scale. The graphs show the mean of two assay repeats and

error bars represent SEM.

024

48

72

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

1 0 1 0

1 0 1 1

1 0 1 2

In tra c e llu la r

T im e (h o u rs )

pfu

/mL R e o (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + D o c (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + D o c (0 .2 5 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + D o c (0 .0 6 × IC 5 0 )

024

48

72

1 0 5

1 0 6

1 0 7

1 0 8

1 0 9

1 0 1 0

1 0 1 1

1 0 1 2

E x tra c e llu la r

T im e (h o u rs )

pfu

/mL R e o (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + D o c (1 .0 0 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + D o c (0 .2 5 × IC 5 0 )

R e o (1 .0 0 × IC 5 0 ) + D o c (0 .0 6 × IC 5 0 )

229

6.3.8 Apoptosis contributes to synergistic cell death when high doses of the

taxane drugs are used in combination with reovirus, but not at low doses

The next step was to assess whether apoptosis or necroptosis was contributing to

synergistic PCa cancer cell death after combination treatment. The DU145 PCa cell

line was treated with reovirus, Cabazitaxel or Docetaxel as single agents at doses

representing 0.25, 0.5, 1.0, 2.0 and 4.0 fold the IC50 values. In addition, cells were

treated with reovirus in combination with Cabazitaxel or Docetaxel at 1.0, 0.50, 0.25,

0.13, 0.06 and 0.03 ×IC50. In each treatment, cells were incubated with the z-VAD-

FMK pan-caspase apoptosis inhibitor (Section 2.25), with the Necrostatin-1 (NCS-1)

necroptosis inhibitor (Section 2.26), or with media alone, for a total of 96-hours. The

cell survival was then determined by the MTS assay (Section 2.9).

Figure 6.16 shows the survival graphs of DU145 cells treated with z-VAD-FMK,

NCS-1 or media alone. In the single agent treated cells, incubation with z-VAD-FMK

caused an increase in cell survival of up to 40% compared to media alone treated

cells, when doses of ≥1.00 ×IC50 were used (p<0.05 by an un-paired t-test). However,

compared to untreated cells, z-VAD-FMK had little effect on cell survival when doses

of <1.00 ×IC50 of each agent was used alone, which also had less statistical power.

Thus, apoptosis may contribute to single agent cytotoxicity at high doses, but not at

low doses. In the combination treated cells, the effect of z-VAD-FMK was visible at

doses between 0.25 and 1.00 ×IC50 (up to 33% cell survival and p<0.05 by an un-

paired t-test), but not at doses <0.25 ×IC50 (insignificant by un-paired t-test), in

comparison to cells treated with media alone. This suggested that apoptosis plays a

role in synergistic anti-cancer cytotoxicity at high doses of reovirus and Cabazitaxel

or Docetaxel, but does not contribute to synergistic cell death at low doses. NCS-1

had an insignificant effect on cell survival in all treatments (p≥0.05), apart from in

cells treated with single-agent reovirus, where there was inhibition of cell death of up

to 14% compared to untreated cells. Hence, it was concluded that necroptosis was not

a major factor involved in synergistic DU145 cell kill.

230

Figure 6.16. High doses of reovirus in combination with Cabazitaxel or Docetaxel causes synergistic anti-cancer cell death by apoptosis, but low dose

combination treatment is non-apoptotic. Cells were treated with A. reovirus, B. Cabazitaxel or C. Docetaxel as single agents at doses representing 0.25, 0.50,

1.00, 2.00 and 4.00 fold the IC50 values. Cells were also treated with D. reovirus in combination with Cabazitaxel or E. reovirus in combination with Docetaxel, at

doses representing 0.03, 0.06, 0.13, 0.25, 0.50 and 1.00 fold the IC50 values. Each treatment condition was incubated with the pan-caspase apoptosis inhibitor z-

VAD-FMK, the necroptosis inhibitor Necrostatin-1 (NCS-1), or media alone. After 96 hours, the % cell survival in each treatment was assessed by the MTS assay,

relative to the untreated cells. z-VAD-FMK caused a significant increase in cell survival compared to cells treated with media alone when high doses of reovirus,

Cabazitaxel or Docetaxel was used alone or in combination. Little effect was observed with NCS-1, indicating that necroptosis is not a major factor that contributes

to cell death. The survival curves display error bars that represent SEM from two assay repeats. *p<0.05, p<0.01 and p<0.001 by un-paired t-test.

0.2

5×

IC5

0

0.5

0×

IC5

0

1.0

0×

IC5

0

2.0

0×

IC5

0

4.0

0×

IC5

0

0

2 5

5 0

7 5

1 0 0

1 2 5


% c

ell

su

rviv

al

***

*

*

0.2

5×

IC5

0

0.5

0×

IC5

0

1.0

0×

IC5

0

2.0

0×

IC5

0

4.0

0×

IC5

0

0

2 5

5 0

7 5

1 0 0

1 2 5


% c

ell

su

rviv

al

****

*

0.2

5×

IC5

0

0.5

0×

IC5

0

1.0

0×

IC5

0

2.0

0×

IC5

0

4.0

0×

IC5

0

0

2 5

5 0

7 5

1 0 0

1 2 5


% c

ell

su

rviv

al

****

**

*

*

0.0

3×

IC5

0

0.0

6×

IC5

0

0.1

3×

IC5

0

0.2

5×

IC5

0

0.5

0×

IC5

0

1.0

0×

IC5

0

0

2 5

5 0

7 5

1 0 0

1 2 5

R e o v iru s (M O I), C a b a z ita x e l (µ M )

% c

ell

su

rviv

al

*******

0.0

3×

IC5

0

0.0

6×

IC5

0

0.1

3×

IC5

0

0.2

5×

IC5

0

0.5

0×

IC5

0

1.0

0×

IC5

0

0

2 5

5 0

7 5

1 0 0

1 2 5

R e o v iru s (M O I) , D o c e ta x e l (µ M )

% c

ell

su

rviv

al

**

**

M e d ia o n ly

z -V A D -F M K

N C S -1

A B C

D E

231

6.4 DISCUSSION

The aim of this chapter was to compare the interaction of reovirus in combination

with either Cabazitaxel or Docetaxel in PCa cell lines, and to explore whether low

doses of each agent in combination could achieve an anti-cancer synergistic effect.

The IC50 values of reovirus, Cabazitaxel or Docetaxel were determined, and then the

level of interaction between the two combinations was assessed by the Chou and

Talalay equation [234]. Bliss Independence analysis was also used to evaluate the

interaction at a range of non-fixed low doses [223, 224]. We assessed whether the

synergistic anti-cancer interaction of the combination treatments was due to enhanced

stabilised microtubules, viral replication, necroptosis or apoptosis.

Different human PCa cell lines have variable molecular and hormonal characteristics.

For example, the LNCaP PCa cell line expresses the androgen receptor, is androgen-

sensitive and expresses PSA [360]. DU145 and PC3 PCa cells however, do not

express the androgen receptor, are androgen-insensitive, and do not produce PSA

[360, 361]. One study showed an inverse relationship between the expression levels

of the androgen receptor and caveolin-1, a plasma-membrane protein that functions to

exchange material between the cell and its environment via endocytosis. LNCaP cells

produced low caveolin-1, whereas DU145 and PC3 cells expressed high caveolin-1

[360]. Interestingly, reovirus ISVPs can exploit caveolar endocytosis to initiate

productive infection [116]. Hence, it would be logical to predict that PCa cells

expressing high levels of caveolin-1 would be more sensitive to reovirus oncolysis

than cells that exhibit low caveolin-1. We found that TRAMP-C2 was the most

susceptible PCa cell line to reovirus oncolysis, followed by DU145, LNCaP and PC3.

Therefore, we did not find a distinct correlation between caveolin-1 expression and

reovirus-induced cell death, as LNCaP were more susceptible to reovirus than PC3.

Normal prostate cells were not available for direct comparison, but interestingly, the

non-cancerous, SV40 transformed WPMY-1 cell line was more resistant to reovirus

than the tumour cell lines. This may be due to the increased chromosomal instability

of the cancer cell lines that could render them more prone to reovirus-induced cell

death [308]. The cell lines, however, showed different sensitivities to the taxane

drugs. Going from the most sensitive to the most resistant cell line, the order of

susceptibility to Cabazitaxel was DU145, LNCaP, TRAMP-C2, WPMY-1 and PC3.

232

Similarly, the order of susceptibility to Docetaxel starting with the most sensitive line

was DU145, TRAMP-C2, LNCaP, WPMY-1 and PC3. Having calculated the IC50

values of each agent, DU145 and LNCaP PCa cell lines were treated concurrently

with the two combinations (reovirus and Cabazitaxel or reovirus and Docetaxel) at

fixed-dose ratios. In both cell lines, the combinations caused a greater synergistic

anti-cancer effect than cells treated with each agent alone, as determined using the

Chou and Talalay equation [234]. The combination of reovirus and Docetaxel was

only fractionally more synergistic than reovirus and Cabazitaxel at ED50, ED75 and

ED90 in DU145 cells. In LNCaP cells, reovirus and Docetaxel showed strong

synergism at the ED50, whereas reovirus and Cabazitaxel showed a slightly lower

level of synergy. Both combinations showed a similar level of synergy at ED75. At

ED90, reovirus and Cabazitaxel was synergistic in LNCaP cells, but reovirus and

Docetaxel showed moderate antagonism. Therefore, the anti-cancer interaction

between reovirus and the taxane drugs at concentrations based around the IC50 values,

varied slightly in different PCa cell lines. The newer taxane entity, Cabazitaxel, has

been shown to maintain activity against docetaxel-resistant cancer cell lines [187,

188]. Thus, we predicted that the combination of reovirus and Cabazitaxel would be

consistently more synergistic than the combination of reovirus and Docetaxel.

However, there was no definitive difference in the anti-cancer effect caused by the

two combination treatments, suggesting that Cabazitaxel and Docetaxel behave

similarly when combined with reovirus at fixed dose ratios around the IC50.

We questioned whether there was an optimal sequence to treat cells with reovirus and

Cabazitaxel or Docetaxel. Clinical trial data has suggested that androgen-deprivation

therapy (ADT) may reduce the efficacy of subsequent taxane treatment for advanced

PCa [185]. This may be because taxanes can inhibit the activity of the androgen

receptor through accumulation of the transcription repressor, forkhead box protein O1

(FOXO1), in the nucleus of PCa cell lines [362]. In turn, this prevents androgen

signalling, which leads to the regression of PCa cancer. Therefore, treatment with

taxane drugs before the administration of ADT may generate a better anti-cancer

response than giving ADT before taxane treatment. We tested five different

sequential combinations of reovirus and Cabazitaxel in the DU145 PCa cell line.

None of the sequential combinations enhanced the synergistic cancer cell kill in

comparison to concurrent combination treatment, suggesting that reovirus does not

233

counteract the activity of taxane drugs, or vice versa. It was therefore decided that

concurrent treatment was the best sequencing strategy. This may influence the way in

which reovirus and taxane drugs are administered in vivo, should this project be taken

further in the future.

Bliss Independence analysis was also used to evaluate the interaction of reovirus and

Cabazitaxel or Docetaxel at a variety of non-fixed doses [223, 224]. The Bliss

Independent model is more flexible than that of Chou and Talalay, and allows the

interaction to be determined at doses below a 50% effect (ED50). Our data suggested

that concentrations of 0.26 ×IC50 reovirus and 0.11 ×IC50 Cabazitaxel in combination,

showed a synergistic anti-cancer effect compared to single agent treatment at these

doses in DU145 cells. In comparison, DU145 cells treated with 0.13 ×IC50 reovirus

and 0.07 ×IC50 Docetaxel in combination generated a synergistic anti-cancer effect

compared to single agent treatment. However, the combination of reovirus and

Cabazitaxel reached a greater level of synergism (46.56%) than the combination of

reovirus and Docetaxel at low doses (29.00%). Therefore, again, it was difficult to

reach a conclusion as to which taxane drug was more efficient when combined with

reovirus at doses below the IC50 values. The data implied that both combinations

have the potential to be used as metronomic treatments for PCa. The original plan

was to test the combinations in the other PCa cell lines, but there was not enough time

left in the project to perform these experiments, which is a possible limitation to the

study.

Since reovirus and taxane drugs have been shown to stabilise cellular microtubules

[141, 192], we predicted that the combination of two microtubule stabilisers would

enhance this effect to promote DU145 cancer cell death. As expected, combination

treatment at the IC50 doses showed greater microtubule stabilisation than single agent

treatment, with reovirus and Cabazitaxel being slightly more efficient than reovirus

and Docetaxel. Both combinations however, enhanced microtubule stability when the

IC50 dose of reovirus was combined with Cabazitaxel or Docetaxel at doses as low as

0.03 ×IC50, compared to treatment with each agent alone. This may partially explain

the synergistic anti-cancer activity after combination treatment at concentrations less

than or equal to the IC50 values.

234

Previous studies have demonstrated that increased viral replication in cancer cells

treated with reovirus and chemotherapeutic agents in combination, correlated with

enhanced cell kill [141, 192]. This has also been shown with oncolytic HSV-1 strains

G207 and NV1066 in combination with various chemotherapy drugs [363-365]. In

contrast to these previously published findings, we found that the intracellular and

extracellular viral titres recovered from DU145 cells treated with the combination of

reovirus and Cabazitaxel or Docetaxel, was generally less than in cells treated with

reovirus alone. This suggested that synergistic PCa cancer cell death caused by

combination treatment is not due to increased viral replication or secretion. As

mentioned in Chapter 5, reovirus yield and cytotoxicity are not closely linked in

some cell types [335, 352]. Mainou et al found that impairment of microtubule

activity by microtubule stabilisers such as taxanes might diminish infection by viruses

that require access to late endosomes to establish a productive infection [359].

Numerous combination studies have shown that chemotherapy agents have no effect

on viral replication [366-369]. In fact, treating malignant pleural mesothelioma cell

lines with the combination of high doses of replication-competent oncolytic HSV-1

and cisplatin, showed strong synergistic cytotoxicity despite a reduction in viral

replication, compared to when lower doses of virus was used in combination [364].

Higher doses of virus resulted in a lower viral titre due to higher loss of cellular

substrates at an earlier time point. Moreover, they showed that loss of cellular

substrates was due to high apoptotic cell fractions, which hindered HSV-1 oncolysis

by limiting viral replication [364]. We hypothesised that a similar effect was taking

place when DU145 cells were treated with reovirus in combination with the taxane

drugs at higher doses. Indeed, we found that cell death caused by high doses of each

agent alone or in combination was partly due to apoptosis, as shown by using a pan-

caspase apoptotic inhibitor (z-VAD-FMK). Apoptosis is a major mechanism of

reovirus-induced cell death, as demonstrated in multiple cancer cell types, including

PCa cell lines [155, 242, 243, 323, 324]. Additionally, taxanes initiate mitotic arrest

and subsequent apoptosis. Firstly, Paclitaxel has been shown to promote apoptosis in

the HeLa cervical adenocarcinoma cell line via activation of cyclin-dependent kinase

1 (Cdk1) (also known as p34cdc2 kinase), a serine/threonine kinase that plays a key

role in cell cycle progression [370]. Secondly, the Bax pro-apoptotic protein

enhanced apoptotic cell death in response to Paclitaxel in ovarian cancer cell lines

[371]. Thirdly, Paclitaxel treatment of PCa cell lines induced the phosphorylation of

235

the anti-apoptotic Bcl-2 protein. This inhibited Bcl-2 binding to Bax, thus disturbing

the balance of pro- and anti- apoptotic interactions, leading to apoptosis [372].

Conversely, we found that z-VAD-FMK had little effect on DU145 cell survival after

low dose treatment of each agent alone or in combination. This somewhat correlated

with the viral titres observed with low dose combination treated cells, which were

more elevated than in high dose combination treated cells. Hence, synergistic cell

death caused by low dose combination treatment may favour an alternative mode of

death such as direct viral replication, as there are more cellular substrates available for

reovirus to replicate. Certain studies have indicated that the necroptosis cell death

pathway can be initiated by reovirus infection [235, 314]. In cells treated with

reovirus alone, there was an increase in cell survival of up to 14% after Necrostatin-1

treatment, indicating that necroptosis may have some involvement in reovirus

oncolysis. However, we were surprised to find that Necrostatin-1 treatment had little

effect on cell survival following the co-administration of reovirus and Cabazitaxel or

Docetaxel, suggesting that necroptosis is not a major contributor of synergistic cell

death in the DU145 PCa cell line.

In addition to the targeting of endothelial cells involved in angiogenesis, metronomic

chemotherapy (MC) can stimulate the immune system to initiate an anti-tumour

response. Regulatory T cells (Tregs) have an immunosuppressive function and

prevent autoimmunity by establishing immunologic self-tolerance [373]. High levels

of Tregs in the tumour microenvironment is associated with poor patient prognosis

[374]. They can supress the anti-cancer immune response by down-regulating the

activity of effector T cells, thus impeding the body’s innate ability to control cancer

cell growth [374]. Several studies have demonstrated that low doses of

cyclophosphamide (an alkylating agent) reduces immunosuppressive CD4+ CD25+

Treg cells, leading to restoration of T and NK cell effector immune functions in end-

stage tumour bearing patients [198], and in mouse models of cancer [199, 200]. MC

cyclophosphamide treatment may also promote the recruitment and maturation of

dendritic cells to the tumour site, leading to tumour regression [201]. Combination

treatment of reovirus and high doses of cyclophosphamide caused wide spread viral

dissemination and cytotoxicity to various organs in C57Bl/6 mice bearing established

sub-cutaneous B16 tumours. However, the metronomic dosing of cyclophosphamide

achieved a balance where the immune response was suppressed enough to allow

236

access of the virus to tumours to enhance viral oncolysis, and caused minimal toxicity

[42, 202]. Treatment of mouse melanoma B16F10 cells with Paclitaxel or Docetaxel

showed enhanced cell-surface expression of calreticulin, a marker of immunogenic

cell death (ICD) and an effective anti-cancer immune response [375]. The

combination of reovirus and an inhibitor of programmed death-1 (PD-1) enhanced the

CD8+ Th1 anti-tumour response in an in vivo melanoma mouse model, resulting in

improved survival, compared to reovirus or anti-PD-1 therapy alone [376]. Therefore,

we predict that the synergistic combination of reovirus and Cabazitaxel or Docetaxel

at low doses may also promote the immune system to help destroy PCa cancer cells.

For future reference, it would be beneficial to test this in culture in cell lines, and in in

vivo PCa mouse models, which may better recapitulate the characteristics of the

disease.

237

6.5 CONCLUSION

The concurrent co-administration of reovirus and Cabazitaxel or Docetaxel at doses

considerably less than the IC50, resulted in a synergistic anti-cancer effect in PCa cell

lines compared to single agent treatment. The biological mechanism of this

interaction may change depending on the dose of the taxane drugs used. The mode

of cell death in low-dose combination treated cells was more influenced by direct

reovirus replication (albeit not to greater levels of single agent reovirus), whereas

high-dose combination treated cells were more likely to be killed by apoptosis.

Microtubule stabilisation was enhanced in both combinations at doses ≤IC50,

suggesting that this is also a factor that contributes to the synergistic interaction. We

conclude that both Cabazitaxel and Docetaxel had equal potential to be used as

metronomic treatments for PCa when combined with oncolytic reovirus. In vivo

investigations into whether MC combination treatment modulates angiogenesis or the

immune system will further support our data.

238

CHAPTER 7

GENERAL DISCUSSION

239

7. GENERAL DISCUSSION

Oncolytic viruses are an attractive treatment option for patients with advanced-stage

malignancies such as SCCHN or PCa, whose prognosis is poor. Despite the

promising development of Reovirus T3D as an anti-tumour agent, there remain

several important areas that warrant further investigation in order to maximise its

oncolytic potential.

The best characterised model of reovirus oncolysis is the ability of reovirus to exploit

cells with a constitutively activated Ras pathway [132-138]; an aberration often

present in cancer cells that facilitates their chronic proliferative signalling [1].

However, a number of reports have argued that reovirus oncolysis can occur

independently of activated Ras and EGFR signalling pathways in different cancers

[139-144], including SCCHN [145]. Hence, the mechanism of reovirus-induced

cancer cell death is still poorly understood, and this may explain the lack of success in

finding a robust predictive biomarker of reovirus treatment response. Biomarkers

have proved to be valuable tools in stratifying the most responsive patient subgroup to

a cancer treatment, and several are operational in the UK [148]. This study addressed

the question of whether a host-cell factor could be identified in a panel of SCCHN

cell lines that may predict for the susceptibility to reovirus oncolysis.

Additionally, there is a growing realisation that reovirus will not display sufficient

efficacy when used as a monotherapy [41, 42, 153]. Of the thirty-six clinical trials

involving Reolysin®, twenty-four have evaluated Reolysin® in combination with

conventional treatments such as chemotherapy or radiation, which have shown to

enhance the tumour cell killing effect [166]. However, administering chemotherapy

drugs at the standard maximum tolerated dose often causes cytotoxicity, and the

extensive drug-free breaks designed to help patients recover can eventually lead to

tumour vasculature re-growth and disease progression [193, 194, 203]. This study

aimed to assess whether the co-administration of reovirus and low, metronomic doses

of chemotherapeutic taxane drugs achieves a synergistic anti-cancer effect in PCa cell

lines, compared to single-agent treatment. This has the potential to limit toxic side-

effects and tumour-associated angiogenesis [195], stimulate an anti-tumour immune

response [198-202], and help sustain clinical responses.

240

7.1. Yes-associated protein 1 (YAP1) as a predictive biomarker of reovirus

treatment response in SCCHN

Preliminary work revealed that the mRNA expression of eight genes increased in a

panel of SCCHN cell lines, as they became progressively more resistant to reovirus

oncolysis, suggesting that they were potential targets of reovirus susceptibility

(R.Morgan, 2007, unpublished) [145] (Section 3.1 and 3.3.1). Having reproduced

and shown congruence with these previous findings in three SCCHN cell lines

(Section 3.3.2 and 3.3.3), this hypothesis was tested by performing knock-down of

the eight genes in the PJ41 cell line. Subsequent infection with reovirus showed that

knock-down of yes-associated protein-1 (YAP1) significantly sensitised the cells to

reovirus, implying that YAP1 may hinder efficient reovirus oncolysis in SCCHN

(Section 3.3.6). Knock-down of the YAP1 protein proved to be less effective than its

gene equivalent (Section 3.3.7), possibly due to the high level of redundancy in the

upstream signalling pathways that control YAP1 [276]. The Hippo signalling

pathway is a major upstream regulator of all YAP isoforms, which shuttle between the

cytoplasm and the nucleus of the cell. Absence of Hippo signalling leads to nuclear

migration of YAP, where it acts as a cofactor to stimulate expression of genes that

promote proliferation [219, 253, 275]. Conversely, core Hippo kinases can

phosphorylate and sequester YAP to the cytoplasm, where it stays inactive [251, 272,

273]. Elevated nuclear YAP is often observed in cancers of the head and neck [377],

lung [378], colon [379], liver [380] and stomach [381], compared to in normal tissues,

and is associated with a poor prognosis [251, 271, 283, 381].

Over-expression of YAP1 and its corresponding protein caused enhanced resistance to

reovirus oncolysis in PJ34 and HN5 cell lines (Section 4.3.2 an 4.3.4). These results

further strengthened the conclusion that the expression level of YAP1 is important in

determining the susceptibility of SCCHN cells to reovirus oncolysis. There appears

to be no evidence linking YAP1 signalling to oncolytic virus modulation, and thus,

our work is the first example. The mechanism behind this was then studied in detail.

It is unlikely that YAP1-mediated restriction of reovirus oncolysis occurs at the cell

surface, as JAM-A receptor expression did not correlate with the SCCHN cell line

reovirus IC50 values, and over-expression of YAP1 did not alter the levels of JAM-A

241

(Section 5.3.2). This is in agreement with the work of Twigger et al and others [144,

145, 348, 349].

The lack of a clear association between viral yield and reovirus oncolysis in PJ34,

HN5 and PJ41, suggested that the mode of killing in these cells is not influenced by

direct reovirus replication, which corresponded with several previously published

reports [145, 335, 350, 352]. Stable over-expression of YAP1 in the HN5 cell line

impeded reovirus replication to some extent, which would partially explain the

increased resistance to oncolysis (Section 5.3.4 and 5.3.5). Artificial over-expression

of YAP1 may behave differently to a cell that expresses similar levels of YAP1, but

endogenously. Differences in the cytoplasmic and nuclear expression levels of YAP1

may play a part in this, and may affect the physiological response to reovirus

infection. A possible limitation to our work was that immunofluorescent staining to

establish YAP1 localisation after YAP1 over-expression or knock-down, was not

performed. Doing so may shed further light on the function of YAP1 in regards to

reovirus oncolysis.

Interferon-β (IFN-β) secretion was not significantly altered by over-expression or

knock-down of YAP1 in SCCHN cell lines after reovirus infection. The level of IFN-

β production correlated with the susceptibility to reovirus oncolysis in PJ34, HN5 and

PJ41 SCCHN cell lines (Section 5.3.7). However, as there were no differences in

reoviral yield between these cell lines, this indicated that they may contain defects

that render them non-responsive to type I interferon, which is a common survival

device employed by tumours [354, 382]. Therefore, YAP1-mediated resistance to

reovirus is probably not linked to the type I interferon anti-viral response.

To test the possibility of using YAP1 as a biomarker, YAP1 protein expression in

head and neck carcinoma tissue was ascertained. 13% of these tumour tissues

expressed YAP1, whereas normal head and neck specimens did not (Section 5.3.8).

Hence, since YAP1 can be measured easily in tumour tissue, that its expression can be

distinguished from normal tissue, and that it is part of a signalling pathway involved

in cancer progression, suggests that YAP1 meets some of the criteria required of a

biomarker of reovirus treatment response in SCCHN. Further validation of the IHC

staining cut-off value and how the tumour tissue would be prepared and processed,

would also be needed. No correlation between YAP1 expression and clinico-

242

pathological data was found, although larger sample numbers would be necessary to

make more reliable conclusions. The next step would be to determine whether there

is an inverse correlation between YAP1 and reovirus protein expression in reovirus-

infected tumours. The utilisation of a predictive biomarker in the clinic may help

target the patients whose tumours are most responsive to reovirus therapy, which

could improve economic factors such as cost and time, as well as strengthening

clinical trial outcome of Reolysin®.

7.2. Combination of reovirus with metronomic doses of taxane drugs in PCa

cell lines

The combination of reovirus and Cabazitaxel, or reovirus and Docetaxel, had an anti-

cancer synergistic effect compared to single agent treatment, at doses considerably

lower than their IC50 values in the DU145 PCa cell line (Section 6.3.4). It was

concluded that the taxane drugs had approximately equal potential to be used as

metronomic treatments for PCa when added concurrently with reovirus. To our

knowledge, this has not been previously explored and therefore, represents a novel

area of research. Given that PCa is a heterogeneous disease and that different human

PCa cell lines have variable molecular and hormonal characteristics, testing the

treatment combinations in more PCa cell lines would help support our data.

An investigation into the mechanism of synergy showed that the mode of DU145 cell

death may change depending on the dose of taxane drug used. Direct reovirus

replication seemed to have more of an effect in cells treated with low-dose

combinations (Section 6.3.7), while apoptotic pathways had greater influence in the

high-dose combination treated cells (Section 6.3.8). A similar effect has been

documented elsewhere in tumour cell lines treated with oncolytic HSV-1 and cisplatin

[364]. Microtubule stabilisation was also a factor that contributed to synergistic

cancer cell kill after combination treatment at doses equal to or less than the IC50

(Section 6.3.6). A disadvantage of this work is that cell lines do not fully take into

account the impact of the immune system. However, our data provides reason to

assess metronomic combination treatment of reovirus and taxane drugs in an in vivo

system, which would be a closer representation of the human body. This may help

improve the way drugs are administered in the clinic, to maximise patient survival.

243

7.3. Future work

The aim of this work was to find ways of enhancing the oncolytic effect of reovirus

via identification of a predictive biomarker of treatment response, and combination

with low, metronomic doses of taxane drugs. Due to time coming to an end, our work

has some unfinished elements that may raise some important questions for new

students wishing to continue this project.

7.3.1. Short-term

We were unable to unearth the complete mechanism of YAP1-mediated resistance to

reovirus oncolysis in SCCHN. However, there are potential factors linking upstream

YAP1 signalling and reovirus infection that would be worth further exploration.

Firstly, YAP1 may be suppressing the efficiency of reovirus-induced apoptosis in

SCCHN cell lines. Nuclear YAP can transcriptionally up-regulate the expression of

anti-apoptotic factors in tumour cell lines [300, 357], and these factors are also known

to inhibit apoptosis induced by reovirus infection [242, 243, 323]. It is possible that a

threshold level of YAP1 might prevent reovirus-induced apoptosis by promoting

expression of anti-apoptotic proteins, hence aiding SCCHN cell survival. A PCR

experiment would verify whether these anti-apoptotic factors are up-regulated after

over-expression of YAP1, or down-regulated after YAP1 knock-down.

Secondly, studies in Drosophila and human cell lines have shown that the actin

cytoskeleton is an upstream regulator of the Hippo pathway [337]. Increased levels of

filamentous (F)-actin in HeLa cells using an actin-stabilising drug resulted in

inhibition of Hippo signalling and caused activation, decreased phosphorylation and

nuclear localisation of YAP, to promote cell growth [383]. Other groups have

concluded that activation of the Hippo pathway and cytoplasmic retention of YAP

decreases the levels of F-actin [384, 385], suggesting that a negative feedback loop

between the Hippo pathway and the actin cytoskeleton may exist [337]. The reoviral

µ2 protein is capable of interacting with and stabilising cellular microtubules to aid

reovirus replication [123, 191, 192]. Thus, perhaps reovirus-induced cell death is

somehow influenced by YAP1 via the dynamic behaviour of cytoskeletal components.

Treating infected cells with cytoskeletal inhibitors such as taxanes, phalloidin or

cytochalasin D, may confirm this theory. Indeed, as microtubule stabilisation is a

244

mode of death by taxane drugs, it would be of value to investigate whether YAP1

signalling influences taxane efficacy in PCa cell lines.

Thirdly, there is evidence suggesting that HPV infection may control upstream

signalling of YAP1 [286]. Since HPV-negative SCCHN cell lines were significantly

more susceptible to reovirus oncolysis than HPV-positive SCCHN cell lines [236], it

may be worth investigating whether HPV status is linked to YAP1 mediated-

resistance to reovirus oncolysis, in PJ34, HN5 and PJ41 cells.

Figure 7.1. Potential mechanism of YAP1-mediated resistance to reovirus oncolysis in SCCHN

cell lines. Reovirus infection causes TRAIL ligands to bind to cell surface death receptors, resulting in

recruitment of FADD and pro-caspase 8. Activated caspase 8 induces cleavage of the pro-apoptotic

protein Bid, which migrates to mitochondria to disrupt the interactions between pro- and anti- apoptotic

Bcl-2 family member proteins. This leads to mitochondrial release of cytochrome c and smac, and

activation of effector caspase 3 for apoptosis. However, inhibition of Hippo pathway signalling after

reovirus infection, possibly via interactions with cytoskeletal components or HPV infection, activates

nuclear YAP1. YAP1 up-regulates anti-apoptotic factors such as Bcl-2 or Bcl-xL, which prevent

mitochondrial release of cytochrome c and smac. Therefore a certain level of YAP1 expression might

prevent apoptosis induced by reovirus.

245

As the effects on reovirus-induced cell death after knock-down or over-expression of

YAP1 were significant, albeit modest, multiple factors must cooperate with YAP1 to

mediate reovirus oncolysis. It would be insightful to confirm which upstream

signalling pathways are acting upon YAP1 in the SCCHN cell lines, and after forced

YAP1 over-expression or knock-down, pre and post reovirus infection. Gene

expression profiling and microarray hybridisation would be a good starting point in

determining the associated up- and down- regulated genes. Concentrating more

specifically on the known signalling activities of YAP1 such as the Hippo pathway

via a PCR array may be a less expensive alternative. Elaborating on the wider

signalling network linking YAP1 with reovirus infection may enhance our

understanding of the oncolysis process, and perhaps a predictive biomarker composed

of several related factors would be more compelling than a single marker on its own.

In vivo experiments would support our data on the metronomic combination of taxane

drugs with oncolytic reovirus. A feasible model would be to treat C57BL/6 mice

bearing TRAMP-C2 tumours, with reovirus intratumourally. The first cohort would

be given frequent intraperitoneal injections of the taxane agent at low doses, whereas

a second cohort would be treated with the taxane at the maximum tolerated dose

(MTD) with longer drug-free periods before the next cycle. Anti-tumour responses

could be compared between the treatment groups, as well as the anti-angiogenic and

immune factors in blood and tumour samples. Initial pilot studies would be needed to

establish the MTD, the minimally effective dose, and the frequency of treatment.

7.3.2. Long-term

Over-expression of YAP1 in COS-1 cells caused significant resistance to reovirus

oncolysis (Section 4.3.6), implying that this may give rise to a cancer-like phenotype,

and that YAP1 may be a universal factor that promotes resistance to reovirus in cells

other than SCCHN. To test this theory, similar experiments to the ones carried out in

this thesis could be applied to different cancer cell types.

Until very recently, the interaction between YAP1 and the androgen receptor (AR)

remained un-explored. Kuser-Abali et al identified YAP1 as a binding partner and a

positive regulator of the AR that therefore plays a critical role in PCa progression

[386]. YAP also plays a role in chemo-resistance; knock-down of the YAP2 isoform

246

sensitised ovarian cancer cell lines to cisplatin treatment [387]. Therefore, it would be

interesting to determine whether YAP1 contributes to reovirus resistance in PCa.

S1P treatment sensitised PJ41 cells to reovirus oncolysis, but we were unable to prove

that this was due to S1P-mediated activation of nuclear YAP1 activity (Section 4.3.8).

Repeating the experiment with longer incubations of S1P before assessing the de-

phosphorylation of YAP1 may confirm this. As there was some nuclear YAP1

present in these cells, determining the effect of a nuclear YAP1 inhibitor on reovirus

oncolysis may also be of interest. Inhibitors of YAP may slow the development of

cancer [291, 386], and in the context of this project, may sensitise SCCHN cell lines

to reovirus. Contemplating this, patients with tumours expressing high levels of

YAP1 could be pre-treated with YAP1 inhibitors as a way of enhancing the oncolytic

effect of reovirus.

In summary, this thesis provides foundations and further questions about optimising

reovirus T3D as an anti-cancer therapeutic.

247

REFERENCES

248

1. Hanahan, D. and R.A. Weinberg, Hallmarks of Cancer: The Next Generation. Cell, 2011. 144(5): p. 646-674.

2. Gump, J.M. and A. Thorburn, Autophagy and apoptosis: what is the connection? Trends Cell Biol, 2011. 21(7): p. 387-92.

3. (NCI), N.C.I. Head and Neck Cancer fact sheet. 2016 1 February 2013. 4. (CRUK), C.R.U. Cancer statistics. 2016 April 2016]; Available from:

http://www.cancerresearchuk.org/health-professional/cancer-statistics/worldwide-cancer. 5. Leemans, C.R., B.J. Braakhuis, and R.H. Brakenhoff, The molecular biology of head and neck

cancer. Nat Rev Cancer, 2011. 11(1): p. 9-22. 6. Chaturvedi, A.K., et al., Incidence trends for human papillomavirus-related and -unrelated

oral squamous cell carcinomas in the United States. J Clin Oncol, 2008. 26(4): p. 612-9. 7. D'Souza, G., et al., Case-control study of human papillomavirus and oropharyngeal cancer. N

Engl J Med, 2007. 356(19): p. 1944-56. 8. Tabor, M.P., et al., Persistence of genetically altered fields in head and neck cancer patients:

biological and clinical implications. Clin Cancer Res, 2001. 7(6): p. 1523-32. 9. Smeets, S.J., et al., Genome-wide DNA copy number alterations in head and neck squamous

cell carcinomas with or without oncogene-expressing human papillomavirus. Oncogene, 2006. 25(17): p. 2558-64.

10. Gillison, M.L., et al., Evidence for a causal association between human papillomavirus and a subset of head and neck cancers. J Natl Cancer Inst, 2000. 92(9): p. 709-20.

11. Hafkamp, H.C., et al., Marked differences in survival rate between smokers and nonsmokers with HPV 16-associated tonsillar carcinomas. Int J Cancer, 2008. 122(12): p. 2656-64.

12. Ang, K.K., et al., Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med, 2010. 363(1): p. 24-35.

13. Reed, A.L., et al., High frequency of p16 (CDKN2/MTS-1/INK4A) inactivation in head and neck squamous cell carcinoma. Cancer Res, 1996. 56(16): p. 3630-3.

14. Ozanne, B., et al., Over-expression of the EGF receptor is a hallmark of squamous cell carcinomas. J Pathol, 1986. 149(1): p. 9-14.

15. Qiu, W., et al., Disruption of transforming growth factor beta-Smad signaling pathway in head and neck squamous cell carcinoma as evidenced by mutations of SMAD2 and SMAD4. Cancer Lett, 2007. 245(1-2): p. 163-70.

16. Murugan, A.K., et al., Oncogenic mutations of the PIK3CA gene in head and neck squamous cell carcinomas. Int J Oncol, 2008. 32(1): p. 101-11.

17. Okami, K., et al., Analysis of PTEN/MMAC1 alterations in aerodigestive tract tumors. Cancer Res, 1998. 58(3): p. 509-11.

18. Iro, H. and F. Waldfahrer, Evaluation of the newly updated TNM classification of head and neck carcinoma with data from 3247 patients. Cancer, 1998. 83(10): p. 2201-7.

19. Macmillan. 1 November 2012 [cited 2014; Available from: http://www.macmillan.org.uk/Cancerinformation/Cancertypes/Headneck/Headneckcancers.aspx.

20. Gourin, C.G. and R.H. Podolsky, Racial disparities in patients with head and neck squamous cell carcinoma. Laryngoscope, 2006. 116(7): p. 1093-106.

21. Malhotra, A., et al., Oncolytic virotherapy for head and neck cancer. Oncolytic Virotherapy, 2015. 4: p. 83-93.

22. (CRUK), C.R.U. Prostate cancer. 2016 May 2016]; Available from: http://www.cancerresearchuk.org/about-cancer/type/prostate-cancer/about/prostate-cancer-types.

23. Struewing, J.P., et al., The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med, 1997. 336(20): p. 1401-8.

24. Ostrander, E.A. and M.S. Udler, The role of the BRCA2 gene in susceptibility to prostate cancer revisited. Cancer Epidemiol Biomarkers Prev, 2008. 17(8): p. 1843-8.

http://www.cancerresearchuk.org/health-professional/cancer-statistics/worldwide-cancer

http://www.macmillan.org.uk/Cancerinformation/Cancertypes/Headneck/Headneckcancers.aspx

http://www.macmillan.org.uk/Cancerinformation/Cancertypes/Headneck/Headneckcancers.aspx

http://www.cancerresearchuk.org/about-cancer/type/prostate-cancer/about/prostate-cancer-types

http://www.cancerresearchuk.org/about-cancer/type/prostate-cancer/about/prostate-cancer-types

249

25. Zeegers, M.P., A. Jellema, and H. Ostrer, Empiric risk of prostate carcinoma for relatives of patients with prostate carcinoma: a meta-analysis. Cancer, 2003. 97(8): p. 1894-903.

26. Chen, Y.C., et al., Family history of prostate and breast cancer and the risk of prostate cancer in the PSA era. Prostate, 2008. 68(14): p. 1582-91.

27. Hoffman, R.M., et al., Racial and ethnic differences in advanced-stage prostate cancer: the Prostate Cancer Outcomes Study. J Natl Cancer Inst, 2001. 93(5): p. 388-95.

28. Renehan, A.G., et al., Insulin-like growth factor (IGF)-I, IGF binding protein-3, and cancer risk: systematic review and meta-regression analysis. Lancet, 2004. 363(9418): p. 1346-53.

29. Rowlands, M.A., et al., Circulating insulin-like growth factor peptides and prostate cancer risk: a systematic review and meta-analysis. Int J Cancer, 2009. 124(10): p. 2416-29.

30. Cogliano, V.J., et al., Preventable exposures associated with human cancers. J Natl Cancer Inst, 2011. 103(24): p. 1827-39.

31. Rodrigues, D.N., et al., Molecular pathology and prostate cancer therapeutics: from biology to bedside. J Pathol, 2014. 232(2): p. 178-84.

32. Tan, M.H., et al., Androgen receptor: structure, role in prostate cancer and drug discovery. Acta Pharmacol Sin, 2015. 36(1): p. 3-23.

33. Traish, A.M. and A. Morgentaler, Epidermal growth factor receptor expression escapes androgen regulation in prostate cancer: a potential molecular switch for tumour growth. Br J Cancer, 2009. 101(12): p. 1949-56.

34. Wu, J.D., et al., Interaction of IGF signaling and the androgen receptor in prostate cancer progression. J Cell Biochem, 2006. 99(2): p. 392-401.

35. Seaton, A., et al., Interleukin-8 signaling promotes androgen-independent proliferation of prostate cancer cells via induction of androgen receptor expression and activation. Carcinogenesis, 2008. 29(6): p. 1148-56.

36. Shariat, S.F., et al., Plasma levels of interleukin-6 and its soluble receptor are associated with prostate cancer progression and metastasis. Urology, 2001. 58(6): p. 1008-15.

37. Nakashima, J., et al., Serum interleukin 6 as a prognostic factor in patients with prostate cancer. Clin Cancer Res, 2000. 6(7): p. 2702-6.

38. Lonergan, P.E. and D.J. Tindall, Androgen receptor signaling in prostate cancer development and progression. J Carcinog, 2011. 10: p. 20.

39. Sarker, D., et al., Targeting the PI3K/AKT pathway for the treatment of prostate cancer. Clin Cancer Res, 2009. 15(15): p. 4799-805.

40. Schurko, B. and W.K. Oh, Docetaxel chemotherapy remains the standard of care in castration-resistant prostate cancer. Nat Clin Pract Oncol, 2008. 5(9): p. 506-7.

41. Hermiston, T.W. and I. Kuhn, Armed therapeutic viruses: strategies and challenges to arming oncolytic viruses with therapeutic genes. Cancer Gene Ther, 2002. 9(12): p. 1022-35.

42. Ottolino-Perry, K., et al., Intelligent design: combination therapy with oncolytic viruses. Mol Ther, 2010. 18(2): p. 251-63.

43. Dock, G., The influence of complicating diseases upon leuk ae mia. American Journal of the Medical Sciences, 1904. 127: p. 563-592.

44. Zamarin, D. and P. Palese, Oncolytic Newcastle disease virus for cancer therapy: old challenges and new directions. Future Microbiol, 2012. 7(3): p. 347-67.

45. Kelly, E. and S.J. Russell, History of Oncolytic Viruses: Genesis to Genetic Engineering. Molecular Therapy, 2007. 15(4): p. 651-659.

46. Martuza, R.L., et al., Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science, 1991. 252(5007): p. 854-6.

47. Dias, J.D., et al., Targeted Chemotherapy for Head and Neck Cancer with a Chimeric Oncolytic Adenovirus Coding for Bifunctional Suicide Protein FCU1. Clinical Cancer Research, 2010. 16(9): p. 2540-2549.

250

48. Reddy, P.S., S. Ganesh, and D.C. Yu, Enhanced gene transfer and oncolysis of head and neck cancer and melanoma cells by fiber chimeric oncolytic adenoviruses. Clin Cancer Res, 2006. 12(9): p. 2869-78.

49. Ganesh, S., et al., Combination therapy with radiation or cisplatin enhances the potency of Ad5/35 chimeric oncolytic adenovirus in a preclinical model of head and neck cancer. Cancer Gene Ther, 2009. 16(5): p. 383-92.

50. Liu, R.Y., et al., An oncolytic adenovirus enhances antiangiogenic and antitumoral effects of a replication-deficient adenovirus encoding endostatin by rescuing its selective replication in nasopharyngeal carcinoma cells. Biochem Biophys Res Commun, 2013. 442(3-4): p. 171-6.

51. Takahashi, H., et al., Telomerase-specific oncolytic adenovirus: antitumor effects on radiation-resistant head and neck squamous cell carcinoma cells. Head Neck, 2014. 36(3): p. 411-8.

52. Huang, P., et al., Direct and distant antitumor effects of a telomerase-selective oncolytic adenoviral agent, OBP-301, in a mouse prostate cancer model. Cancer Gene Ther, 2008. 15(5): p. 315-22.

53. Hu, Z., et al., Systemic delivery of oncolytic adenoviruses targeting transforming growth factor-beta inhibits established bone metastasis in a prostate cancer mouse model. Hum Gene Ther, 2012. 23(8): p. 871-82.

54. Calvo, E., et al., A Phase 1 Study of Enadenotucirev, an Oncolytic Ad11/Ad3 Chimeric Group B Adenovirus, Administered Intravenously - Analysis of Dose Expansion and Repeat Cycle Cohorts in Patients with Metastatic Colorectal Cancer (Mcrc). Annals of Oncology, 2014. 25.

55. Nemunaitis, J., et al., Phase II trial of intratumoral administration of ONYX-015, a replication-selective adenovirus, in patients with refractory head and neck cancer. J Clin Oncol, 2001. 19(2): p. 289-98.

56. Jiang, H., et al., Oncolytic adenovirus: preclinical and clinical studies in patients with human malignant gliomas. Curr Gene Ther, 2009. 9(5): p. 422-7.

57. Nemunaitis, J., et al., Selective replication and oncolysis in p53 mutant tumors with ONYX-015, an E1B-55kD gene-deleted adenovirus, in patients with advanced head and neck cancer: a phase II trial. Cancer Res, 2000. 60(22): p. 6359-66.

58. release, A.p., Amgen announces top-line results of phase 3 talimogene laherparepvec trial in melanoma. 2013.

59. (CRUK), C.R.U. A trial of T-VEC (Talimogene laherparepvec) for melanoma that cannot be removed. 2016 May 2016]; Available from: http://www.cancerresearchuk.org/about-cancer/find-a-clinical-trial/trial-oncovex-gmcsf-melanoma-cannot-be-removed#undefined.

60. Harrington, K.J., et al., Phase I/II study of oncolytic HSV GM-CSF in combination with radiotherapy and cisplatin in untreated stage III/IV squamous cell cancer of the head and neck. Clin Cancer Res, 2010. 16(15): p. 4005-15.

61. Fujimoto, Y., et al., Intratumoral injection of herpes simplex virus HF10 in recurrent head and neck squamous cell carcinoma. Acta Otolaryngol, 2006. 126(10): p. 1115-7.

62. Ogawa, F., et al., Combined oncolytic virotherapy with herpes simplex virus for oral squamous cell carcinoma. Anticancer Res, 2008. 28(6A): p. 3637-45.

63. Ahmed, M., Oncolytic Viruses as Therapeutic Agents for Prostate Cancer. Adv Tech Biol Med 2013. 1(2).

64. Smith, G.L. and M. Law, The exit of vaccinia virus from infected cells. Virus Res, 2004. 106(2): p. 189-97.

65. Yu, Z.J., et al., Oncolytic vaccinia therapy of squamous cell carcinoma. Molecular Cancer, 2009. 8.

66. Yu, Y.A., et al., Regression of human pancreatic tumor xenografts in mice after a single systemic injection of recombinant vaccinia virus GLV-1h68. Molecular Cancer Therapeutics, 2009. 8(1): p. 141-151.

http://www.cancerresearchuk.org/about-cancer/find-a-clinical-trial/trial-oncovex-gmcsf-melanoma-cannot-be-removed#undefined

http://www.cancerresearchuk.org/about-cancer/find-a-clinical-trial/trial-oncovex-gmcsf-melanoma-cannot-be-removed#undefined

251

67. Gentschev, I., et al., Regression of human prostate tumors and metastases in nude mice following treatment with the recombinant oncolytic vaccinia virus GLV-1h68. J Biomed Biotechnol, 2010. 2010: p. 489759.

68. Eder, J.P., et al., A phase I trial of a recombinant vaccinia virus expressing prostate-specific antigen in advanced prostate cancer. Clin Cancer Res, 2000. 6(5): p. 1632-8.

69. Health, U.S.N.I.o. Hepatocellular Carcinoma Study Comparing Vaccinia Virus Based Immunotherapy Plus Sorafenib vs Sorafenib Alone (PHOCUS). 2016 May 2016]; Available from: https://clinicaltrials.gov/ct2/show/NCT02562755?term=jx594&rank=15.

70. Goldufsky, J., et al., Oncolytic virus therapy for cancer. Oncolytic Virotherapy, 2013. 2: p. 31-46

71. Ockert, D., et al., Newcastle disease virus-infected intact autologous tumor cell vaccine for adjuvant active specific immunotherapy of resected colorectal carcinoma. Clin Cancer Res, 1996. 2(1): p. 21-8.

72. Steiner, H.H., et al., Antitumor vaccination of patients with glioblastoma multiforme: a pilot study to assess feasibility, safety, and clinical benefit. J Clin Oncol, 2004. 22(21): p. 4272-81.

73. Karcher, J., et al., Antitumor vaccination in patients with head and neck squamous cell carcinomas with autologous virus-modified tumor cells. Cancer Res, 2004. 64(21): p. 8057-61.

74. Vigil, A., et al., Use of reverse genetics to enhance the oncolytic properties of newcastle disease virus. Cancer Research, 2007. 67(17): p. 8285-8292.

75. Li, P., et al., Therapeutic effects of a fusogenic newcastle disease virus in treating head and neck cancer. Head Neck, 2011. 33(10): p. 1394-9.

76. Alkassar, M., et al., The combined effects of oncolytic reovirus plus Newcastle disease virus and reovirus plus parvovirus on U87 and U373 cells in vitro and in vivo. J Neurooncol, 2011. 104(3): p. 715-27.

77. Ayala-Breton, C., et al., Retargeting Vesicular Stomatitis Virus Using Measles Virus Envelope Glycoproteins. Human Gene Therapy, 2012. 23(5): p. 484-491.

78. Health, U.S.N.I.o. Viral Therapy in Treating Patient With Liver Cancer. 2016 May 2016]; Available from: https://clinicaltrials.gov/ct2/show/NCT01628640.

79. Miyamoto, S., et al., Coxsackievirus B3 is an oncolytic virus with immunostimulatory properties that is active against lung adenocarcinoma. Cancer Res, 2012. 72(10): p. 2609-21.

80. Shafren, D.R., et al., Coxsackievirus A21 binds to decay-accelerating factor but requires intercellular adhesion molecule 1 for cell entry. J Virol, 1997. 71(6): p. 4736-43.

81. Au, G.G., et al., Oncolytic Coxsackievirus A21 as a novel therapy for multiple myeloma. Br J Haematol, 2007. 137(2): p. 133-41.

82. Berry, L.J., et al., Potent oncolytic activity of human enteroviruses against human prostate cancer. Prostate, 2008. 68(6): p. 577-87.

83. Skelding, K.A., R.D. Barry, and D.R. Shafren, Systemic targeting of metastatic human breast tumor xenografts by Coxsackievirus A21. Breast Cancer Res Treat, 2009. 113(1): p. 21-30.

84. Shafren, D.R., et al., Systemic therapy of malignant human melanoma tumors by a common cold-producing enterovirus, coxsackievirus a21. Clin Cancer Res, 2004. 10(1 Pt 1): p. 53-60.

85. Andtbacka, R.H., et al., CALM Study: A Phase II Study of a Novel Oncolytic Immunotherapeutic Agent, CVA21, Delivered Intratumorally in Patients with Advanced Malignant Melanoma. Annals of Surgical Oncology, 2015. 22: p. S22-S22.

86. Andtbacka, R.H.I., et al., Final data from CALM: A phase II study of Coxsackievirus A21 (CVA21) oncolytic virus immunotherapy in patients with advanced melanoma. Journal of Clinical Oncology, 2015. 33(15).

87. Castelo-Branco, P., et al., Oncolytic herpes simplex virus armed with xenogeneic homologue of prostatic acid phosphatase enhances antitumor efficacy in prostate cancer. Gene Therapy, 2010. 17(6): p. 805-810.

https://clinicaltrials.gov/ct2/show/NCT02562755?term=jx594&rank=15

https://clinicaltrials.gov/ct2/show/NCT01628640

252

88. Passer, B.J., et al., Combination of vinblastine and oncolytic herpes simplex virus vector expressing IL-12 therapy increases antitumor and antiangiogenic effects in prostate cancer models. Cancer Gene Ther, 2013. 20(1): p. 17-24.

89. Conner, J. and L. Braidwood, Expression of inhibitor of growth 4 by HSV1716 improves oncolytic potency and enhances efficacy. Cancer Gene Therapy, 2012. 19(7): p. 499-507.

90. Yuan, Z.Y., et al., [Safety of an E1B deleted adenovirus administered intratumorally to patients with cancer]. Ai Zheng, 2003. 22(3): p. 310-3.

91. Xia, Z.J., et al., [Phase III randomized clinical trial of intratumoral injection of E1B gene-deleted adenovirus (H101) combined with cisplatin-based chemotherapy in treating squamous cell cancer of head and neck or esophagus]. Ai Zheng, 2004. 23(12): p. 1666-70.

92. Xu, R.H., et al., [Phase II clinical study of intratumoral H101, an E1B deleted adenovirus, in combination with chemotherapy in patients with cancer]. Ai Zheng, 2003. 22(12): p. 1307-10.

93. Hu, J.C., et al., A phase I study of OncoVEXGM-CSF, a second-generation oncolytic herpes simplex virus expressing granulocyte macrophage colony-stimulating factor. Clin Cancer Res, 2006. 12(22): p. 6737-47.

94. Ferris, R.L., et al., Phase I trial of intratumoral therapy using HF10, an oncolytic HSV-1, demonstrates safety in HSV plus /HSV- patients with refractory and superficial cancers. Journal of Clinical Oncology, 2014. 32(15).

95. Park, B.H., et al., Use of a targeted oncolytic poxvirus, JX-594, in patients with refractory primary or metastatic liver cancer: a phase I trial. Lancet Oncol, 2008. 9(6): p. 533-42.

96. Park, S.H., et al., Phase 1b Trial of Biweekly Intravenous Pexa-Vec (JX-594), an Oncolytic and Immunotherapeutic Vaccinia Virus in Colorectal Cancer. Mol Ther, 2015. 23(9): p. 1532-40.

97. Cripe, T.P., et al., Phase 1 study of intratumoral Pexa-Vec (JX-594), an oncolytic and immunotherapeutic vaccinia virus, in pediatric cancer patients. Mol Ther, 2015. 23(3): p. 602-8.

98. Sahin, E.E., M.; McMasters, K.; Zhou, H. , Development of Oncolytic Reovirus for Cancer Therapy. Journal of Cancer Therapy, 2013. 4(6): p. 1100-1115.

99. Fields, B.N., D.M. Knipe, and P.M. Howley, Fields virology. 3rd ed. 1996, Philadelphia: Lippincott-Raven Publishers.

100. Rosen, L., H.E. Evans, and A. Spickard, Reovirus infections in human volunteers. Am J Hyg, 1963. 77: p. 29-37.

101. Selb, B. and B. Weber, A study of human reovirus IgG and IgA antibodies by ELISA and western blot. J Virol Methods, 1994. 47(1-2): p. 15-25.

102. Harrington, K.J., H. Pandha, and R.G. Vile, Viral therapy of cancer. 2008, Chichester, England ; Hoboken, NJ: John Wiley & Sons. xx, 404 p.

103. Brubaker, M.M., B. West, and R.J. Eillis, Human Blood Group Influence on Reovirus Hemagglutination Titers. Proc Soc Exp Biol Med, 1964. 115: p. 1118-20.

104. Eggers, H.J., P.J. Gomatos, and I. Tamm, Agglutination of bovine erythrocytes: a general characteristic of reovirus type 3. Proc Soc Exp Biol Med, 1962. 110: p. 879-81.

105. Weiner, H.L., et al., Identification of the gene coding for the hemagglutinin of reovirus. Virology, 1978. 86(2): p. 581-4.

106. Gentsch, J.R. and A.F. Pacitti, Effect of neuraminidase treatment of cells and effect of soluble glycoproteins on type 3 reovirus attachment to murine L cells. J Virol, 1985. 56(2): p. 356-64.

107. Chappell, J.D., et al., Mutations in type 3 reovirus that determine binding to sialic acid are contained in the fibrous tail domain of viral attachment protein sigma 1. Journal of Virology, 1997. 71(3): p. 1834-1841.

108. Chappell, J.D., et al., Identification of carbohydrate-binding domains in the attachment proteins of type 1 and type 3 reoviruses. Journal of Virology, 2000. 74(18): p. 8472-9.

109. Barton, E.S., et al., Junction adhesion molecule is a receptor for reovirus. Cell, 2001. 104(3): p. 441-451.

253

110. Barton, E.S., et al., Utilization of sialic acid as a coreceptor is required for reovirus-induced biliary disease. J Clin Invest, 2003. 111(12): p. 1823-33.

111. Maginnis, M.S., et al., Beta1 integrin mediates internalization of mammalian reovirus. J Virol, 2006. 80(6): p. 2760-70.

112. Ehrlich, M., et al., Endocytosis by random initiation and stabilization of clathrin-coated pits. Cell, 2004. 118(5): p. 591-605.

113. Ebert, D.H., et al., Cathepsin L and cathepsin B mediate reovirus disassembly in murine fibroblast cells. J Biol Chem, 2002. 277(27): p. 24609-17.

114. Golden, J.W., et al., Cathepsin S supports acid-independent infection by some reoviruses. J Biol Chem, 2004. 279(10): p. 8547-57.

115. Dichter, M.A. and H.L. Weiner, Infection of neuronal cell cultures with reovirus mimics in vitro patterns of neurotropism. Ann Neurol, 1984. 16(5): p. 603-10.

116. Schulz, W.L., A.K. Haj, and L.A. Schiff, Reovirus uses multiple endocytic pathways for cell entry. J Virol, 2012. 86(23): p. 12665-75.

117. Borsa, J., et al., 2 Modes of Entry of Reovirus Particles into L-Cells. Journal of General Virology, 1979. 45(Oct): p. 161-170.

118. Zhang, L., et al., Reovirus mu1 structural rearrangements that mediate membrane penetration. J Virol, 2006. 80(24): p. 12367-76.

119. Ivanovic, T., et al., Peptides released from reovirus outer capsid form membrane pores that recruit virus particles. EMBO J, 2008. 27(8): p. 1289-98.

120. Nibert, M.L., et al., Putative autocleavage of reovirus mu1 protein in concert with outer-capsid disassembly and activation for membrane permeabilization. J Mol Biol, 2005. 345(3): p. 461-74.

121. Tosteson, M.T., M.L. Nibert, and B.N. Fields, Ion channels induced in lipid bilayers by subvirion particles of the nonenveloped mammalian reoviruses. Proc Natl Acad Sci U S A, 1993. 90(22): p. 10549-52.

122. Schonberg, M., et al., Asynchronous synthesis of the complementary strands of the reovirus genome. Proc Natl Acad Sci U S A, 1971. 68(2): p. 505-8.

123. Parker, J.S., et al., Reovirus core protein mu2 determines the filamentous morphology of viral inclusion bodies by interacting with and stabilizing microtubules. J Virol, 2002. 76(9): p. 4483-96.

124. Hashiro, G., P.C. Loh, and J.T. Yau, The preferential cytotoxicity of reovirus for certain transformed cell lines. Arch Virol, 1977. 54(4): p. 307-15.

125. Duncan, M.R., S.M. Stanish, and D.C. Cox, Differential sensitivity of normal and transformed human cells to reovirus infection. J Virol, 1978. 28(2): p. 444-9.

126. Douville, R.N., et al., Reovirus serotypes elicit distinctive patterns of recall immunity in humans. J Virol, 2008. 82(15): p. 7515-23.

127. Alloussi, S.H., et al., All reovirus subtypes show oncolytic potential in primary cells of human high-grade glioma. Oncol Rep, 2011. 26(3): p. 645-9.

128. Campbell, P.M. and C.J. Der, Oncogenic Ras and its role in tumor cell invasion and metastasis. Seminars in Cancer Biology, 2004. 14(2): p. 105-114.

129. Repasky, G.A., E.J. Chenette, and C.J. Der, Renewing the conspiracy theory debate: does Raf function alone to mediate Ras oncogenesis? Trends Cell Biol, 2004. 14(11): p. 639-47.

130. Bos, J.L., Ras Oncogenes in Human Cancer - a Review. Cancer Research, 1989. 49(17): p. 4682-4689.

131. Shmulevitz, M., P. Marcato, and P.W.K. Lee, Unshackling the links between reovirus oncolysis, Ras signaling, translational control and cancer. Oncogene, 2005. 24(52): p. 7720-7728.

132. Strong, J.E., D.M. Tang, and P.W.K. Lee, Evidence That the Epidermal Growth-Factor Receptor on Host-Cells Confers Reovirus Infection Efficiency. Virology, 1993. 197(1): p. 405-411.

254

133. Strong, J.E. and P.W. Lee, The v-erbB oncogene confers enhanced cellular susceptibility to reovirus infection. J Virol, 1996. 70(1): p. 612-6.

134. Norman, K.L., et al., Reovirus oncolysis: the Ras/RalGEF/p38 pathway dictates host cell permissiveness to reovirus infection. Proc Natl Acad Sci U S A, 2004. 101(30): p. 11099-104.

135. Strong, J.E., et al., The molecular basis of viral oncolysis: usurpation of the Ras signaling pathway by reovirus. EMBO J, 1998. 17(12): p. 3351-62.

136. Etoh, T., et al., Oncolytic viral therapy for human pancreatic cancer cells by reovirus. Clin Cancer Res, 2003. 9(3): p. 1218-23.

137. Wilcox, M.E., et al., Reovirus as an oncolytic agent against experimental human malignant gliomas. Journal of the National Cancer Institute, 2001. 93(12): p. 903-912.

138. Hirasawa, K., et al., Oncolytic reovirus against ovarian and colon cancer. Cancer Res, 2002. 62(6): p. 1696-701.

139. Song, L., et al., Reovirus infection of cancer cells is not due to activated Ras pathway. Cancer Gene Ther, 2009. 16(4): p. 382.

140. van Houdt, W.J., et al., Transient infection of freshly isolated human colorectal tumor cells by reovirus T3D intermediate subviral particles. Cancer Gene Therapy, 2008. 15(5): p. 284-292.

141. Sei, S., et al., Synergistic antitumor activity of oncolytic reovirus and chemotherapeutic agents in non-small cell lung cancer cells. Mol Cancer, 2009. 8: p. 47.

142. Alain, T., et al., Proteolytic disassembly is a critical determinant for reovirus oncolysis. Mol Ther, 2007. 15(8): p. 1512-21.

143. Golden, J.W., et al., Addition of exogenous protease facilitates reovirus infection in many restrictive cells. J Virol, 2002. 76(15): p. 7430-43.

144. Terasawa, Y., et al., Activity levels of cathepsins B and L in tumor cells are a biomarker for efficacy of reovirus-mediated tumor cell killing. Cancer Gene Ther, 2015. 22(4): p. 188-97.

145. Twigger, K., et al., Reovirus exerts potent oncolytic effects in head and neck cancer cell lines that are independent of signalling in the EGFR pathway. Bmc Cancer, 2012. 12.

146. La Thangue, N.B. and D.J. Kerr, Predictive biomarkers: a paradigm shift towards personalized cancer medicine. Nat Rev Clin Oncol, 2011. 8(10): p. 587-96.

147. Institute, N.C. Biomarker. 2016 10-05-16]; Available from: http://www.cancer.gov/publications/dictionaries/cancer-terms?cdrid=45618.

148. Institute, N.C. Tumor Markers. 2016 10-05-16]; Available from: http://www.cancer.gov/about-cancer/diagnosis-staging/diagnosis/tumor-markers-fact-sheet.

149. Henry, N.L. and D.F. Hayes, Cancer biomarkers. Mol Oncol, 2012. 6(2): p. 140-6. 150. Flak, M.B., et al., p21 Promotes oncolytic adenoviral activity in ovarian cancer and is a

potential biomarker. Mol Cancer, 2010. 9: p. 175. 151. Zloza, A., et al., Immunoglobulin-like transcript 2 (ILT2) is a biomarker of therapeutic

response to oncolytic immunotherapy with vaccinia viruses. J Immunother Cancer, 2014. 2: p. 1.

152. Thirukkumaran, C. and D.G. Morris, Oncolytic Viral Therapy Using Reovirus. Gene Therapy of Solid Cancers: Methods and Protocols, 2015. 1317: p. 187-223.

153. Black, A.J. and D.G. Morris, Clinical trials involving the oncolytic virus, reovirus: ready for prime time? Expert Rev Clin Pharmacol, 2012. 5(5): p. 517-20.

154. Norman, K.L., et al., Reovirus oncolysis of human breast cancer. Hum Gene Ther, 2002. 13(5): p. 641-52.

155. Thirukkumaran, C.M., et al., Oncolytic viral therapy for prostate cancer: efficacy of reovirus as a biological therapeutic. Cancer Res, 2010. 70(6): p. 2435-44.

156. Hanel, E.G., et al., A novel intravesical therapy for superficial bladder cancer in an orthotopic model: oncolytic reovirus therapy. J Urol, 2004. 172(5 Pt 1): p. 2018-22.

157. Hirasawa, K., et al., Systemic reovirus therapy of metastatic cancer in immune-competent mice. Cancer Res, 2003. 63(2): p. 348-53.

http://www.cancer.gov/publications/dictionaries/cancer-terms?cdrid=45618

http://www.cancer.gov/about-cancer/diagnosis-staging/diagnosis/tumor-markers-fact-sheet

http://www.cancer.gov/about-cancer/diagnosis-staging/diagnosis/tumor-markers-fact-sheet

255

158. Alain, T., et al., Reovirus therapy of lymphoid malignancies. Blood, 2002. 100(12): p. 4146-53. 159. Thirukkumaran, C.M., et al., Reovirus oncolysis as a novel purging strategy for autologous

stem cell transplantation. Blood, 2003. 102(1): p. 377-87. 160. Liang, K., et al., The epidermal growth factor receptor mediates radioresistance. Int J Radiat

Oncol Biol Phys, 2003. 57(1): p. 246-54. 161. Bernhard, E.J., et al., The farnesyltransferase inhibitor FTI-277 radiosensitizes H-ras-

transformed rat embryo fibroblasts. Cancer Res, 1996. 56(8): p. 1727-30. 162. McKenna, W.G., et al., The RAS signal transduction pathway and its role in radiation

sensitivity. Oncogene, 2003. 22(37): p. 5866-75. 163. Gupta, A.K., et al., Radiation sensitization of human cancer cells in vivo by inhibiting the

activity of PI3K using LY294002. Int J Radiat Oncol Biol Phys, 2003. 56(3): p. 846-53. 164. Twigger, K., et al., Enhanced in vitro and in vivo cytotoxicity of combined reovirus and

radiotherapy. Clin Cancer Res, 2008. 14(3): p. 912-23. 165. Pandha, H.S., et al., Synergistic effects of oncolytic reovirus and cisplatin chemotherapy in

murine malignant melanoma. Clin Cancer Res, 2009. 15(19): p. 6158-66. 166. Inc, O.B. REOLYSIN® Clinical trials. 2016 10-05-16]; Available from:

http://www.oncolyticsbiotech.com/reolysin/clinical-trials/. 167. Morris, D.G., et al., REO-001: A phase I trial of percutaneous intralesional administration of

reovirus type 3 dearing (Reolysin(R)) in patients with advanced solid tumors. Invest New Drugs, 2013. 31(3): p. 696-706.

168. Forsyth, P., et al., A phase I trial of intratumoral administration of reovirus in patients with histologically confirmed recurrent malignant gliomas. Mol Ther, 2008. 16(3): p. 627-32.

169. Vidal, L., et al., A phase I study of intravenous oncolytic reovirus type 3 Dearing in patients with advanced cancer. Clin Cancer Res, 2008. 14(21): p. 7127-37.

170. White, C.L., et al., Characterization of the adaptive and innate immune response to intravenous oncolytic reovirus (Dearing type 3) during a phase I clinical trial. Gene Ther, 2008. 15(12): p. 911-20.

171. Harrington, K.J., et al., Two-stage phase I dose-escalation study of intratumoral reovirus type 3 dearing and palliative radiotherapy in patients with advanced cancers. Clin Cancer Res, 2010. 16(11): p. 3067-77.

172. Saunders, M., et al., Results of a phase II study to evaluate the biological effects of intratumoral (ITu) reolysin in combination with low dose radiotherapy (RT) in patients (Pts) with advanced cancers. Journal of Clinical Oncology, 2009. 27(15).

173. Lolkema, M.P., et al., A phase I study of the combination of intravenous reovirus type 3 Dearing and gemcitabine in patients with advanced cancer. Clin Cancer Res, 2011. 17(3): p. 581-8.

174. Comins, C., et al., REO-10: a phase I study of intravenous reovirus and docetaxel in patients with advanced cancer. Clin Cancer Res, 2010. 16(22): p. 5564-72.

175. Karapanagiotou, E.M., et al., Phase I/II trial of carboplatin and paclitaxel chemotherapy in combination with intravenous oncolytic reovirus in patients with advanced malignancies. Clin Cancer Res, 2012. 18(7): p. 2080-9.

176. Roulstone, V., et al., Phase I Trial of Cyclophosphamide as an Immune Modulator for Optimizing Oncolytic Reovirus Delivery to Solid Tumors. Clinical Cancer Research, 2015. 21(6): p. 1305-1312.

177. Cohn, D.E., et al., Phase I/II trial of reovirus serotype 3-Dearing strain in patients with recurrent ovarian cancer. Journal of Clinical Oncology, 2010. 28(15).

178. Galanis, E., et al., Phase II trial of intravenous administration of Reolysin((R)) (Reovirus Serotype-3-dearing Strain) in patients with metastatic melanoma. Mol Ther, 2012. 20(10): p. 1998-2003.

179. Adair, R.A., et al., Cell carriage, delivery, and selective replication of an oncolytic virus in tumor in patients. Sci Transl Med, 2012. 4(138): p. 138ra77.

http://www.oncolyticsbiotech.com/reolysin/clinical-trials/

256

180. Kolb, E.A., et al., A Phase I Trial and Viral Clearance Study of Reovirus (Reolysin) in Children With Relapsed or Refractory Extracranial Solid Tumors: A Children's Oncology Group Phase I Consortium Report. Pediatric Blood & Cancer, 2015. 62(5): p. 751-758.

181. Sborov, D.W., et al., A phase I trial of single-agent reolysin in patients with relapsed multiple myeloma. Clin Cancer Res, 2014. 20(23): p. 5946-55.

182. Galsky, M.D., et al., Cabazitaxel. Nat Rev Drug Discov, 2010. 9(9): p. 677-8. 183. Tsao, C.K., et al., Clinical development of cabazitaxel for the treatment of castration-resistant

prostate cancer. Clin Med Insights Oncol, 2011. 5: p. 163-9. 184. Riou, J.F., A. Naudin, and F. Lavelle, Effects of Taxotere on murine and human tumor cell

lines. Biochem Biophys Res Commun, 1992. 187(1): p. 164-70. 185. Fitzpatrick, J.M. and R. de Wit, Taxane mechanisms of action: potential implications for

treatment sequencing in metastatic castration-resistant prostate cancer. Eur Urol, 2014. 65(6): p. 1198-204.

186. Abidi, A., Cabazitaxel: A novel taxane for metastatic castration-resistant prostate cancer-current implications and future prospects. J Pharmacol Pharmacother, 2013. 4(4): p. 230-237.

187. Mita, A.C., et al., Phase I and pharmacokinetic study of XRP6258 (RPR 116258A), a novel taxane, administered as a 1-hour infusion every 3 weeks in patients with advanced solid tumors. Clin Cancer Res, 2009. 15(2): p. 723-30.

188. Sanofi-Aventis, XRP6258 investigator's brochure. . 2000. 189. Babiss, L.E., et al., Reovirus serotypes 1 and 3 differ in their in vitro association with

microtubules. J Virol, 1979. 30(3): p. 863-74. 190. Kim, J., et al., Nucleoside and RNA triphosphatase activities of orthoreovirus transcriptase

cofactor mu2. J Biol Chem, 2004. 279(6): p. 4394-403. 191. Kobayashi, T., et al., Identification of functional domains in reovirus replication proteins

muNS and mu2. J Virol, 2009. 83(7): p. 2892-906. 192. Heinemann, L., et al., Synergistic effects of oncolytic reovirus and docetaxel chemotherapy in

prostate cancer. BMC Cancer, 2011. 11: p. 221. 193. Browder, T., et al., Antiangiogenic scheduling of chemotherapy improves efficacy against

experimental drug-resistant cancer. Cancer Res, 2000. 60(7): p. 1878-86. 194. Faivre, C., et al., A mathematical model for the administration of temozolomide: comparative

analysis of conventional and metronomic chemotherapy regimens. Cancer Chemother Pharmacol, 2013. 71(4): p. 1013-9.

195. Scharovsky, O.G., L.E. Mainetti, and V.R. Rozados, Metronomic chemotherapy: changing the paradigm that more is better. Curr Oncol, 2009. 16(2): p. 7-15.

196. Zou, L., et al., Lasting controversy on ranibizumab and bevacizumab. Theranostics, 2011. 1: p. 395-402.

197. Pasquier, E., M. Kavallaris, and N. Andre, Metronomic chemotherapy: new rationale for new directions. Nat Rev Clin Oncol, 2010. 7(8): p. 455-65.

198. Ghiringhelli, F., et al., Metronomic cyclophosphamide regimen selectively depletes CD4+CD25+ regulatory T cells and restores T and NK effector functions in end stage cancer patients. Cancer Immunol Immunother, 2007. 56(5): p. 641-8.

199. Lutsiak, M.E., et al., Inhibition of CD4(+)25+ T regulatory cell function implicated in enhanced immune response by low-dose cyclophosphamide. Blood, 2005. 105(7): p. 2862-8.

200. Loeffler, M., J.A. Kruger, and R.A. Reisfeld, Immunostimulatory effects of low-dose cyclophosphamide are controlled by inducible nitric oxide synthase. Cancer Res, 2005. 65(12): p. 5027-30.

201. Doloff, J.C. and D.J. Waxman, VEGF receptor inhibitors block the ability of metronomically dosed cyclophosphamide to activate innate immunity-induced tumor regression. Cancer Res, 2012. 72(5): p. 1103-15.

202. Qiao, J., et al., Cyclophosphamide facilitates antitumor efficacy against subcutaneous tumors following intravenous delivery of reovirus. Clin Cancer Res, 2008. 14(1): p. 259-69.

257

203. Klement, G.L. and B.A. Kamen, Nontoxic, Fiscally Responsible, Future of Oncology: Could it be Beginning in the Third World? Journal of Pediatric Hematology Oncology, 2011. 33(1): p. 1-3.

204. Wang, Z.H., et al., An all-oral combination of metronomic cyclophosphamide plus capecitabine in patients with anthracycline- and taxane-pretreated metastatic breast cancer: a phase II study. Cancer Chemotherapy and Pharmacology, 2012. 69(2): p. 515-522.

205. Garcia, J.L.S., et al., Metronomic Cyclophosphamide and Methotrexate Chemotherapy Combined with 1e10 Anti-Idiotype Vaccine in Metastatic Breast Cancer. Annals of Oncology, 2010. 21: p. 182-183.

206. Borne, E., et al., Oral metronomic cyclophosphamide in elderly with metastatic melanoma. Investigational New Drugs, 2010. 28(5): p. 684-689.

207. Gunsilius, E., et al., Palliative chemotherapy in pretreated patients with advanced cancer: Oral trofosfamide is effective in ovarian carcinoma. Cancer Investigation, 2001. 19(8): p. 808-811.

208. Herrlinger, U., et al., UKT-04 trial of continuous metronomic low-dose chemotherapy with methotrexate and cyclophosphamide for recurrent glioblastoma. Journal of Neuro-Oncology, 2005. 71(3): p. 295-299.

209. Lord, R., et al., Low dose metronomic oral cyclophosphamide for hormone resistant prostate cancer: A phase II study. Journal of Urology, 2007. 177(6): p. 2136-2140.

210. Correale, P., et al., A novel metronomic chemotherapy regimen of weekly platinum and daily oral etoposide in high-risk non-small cell lung cancer patients. Oncology Reports, 2006. 16(1): p. 133-140.

211. Bocci, G., K.C. Nicolaou, and R.S. Kerbel, Protracted low-dose effects on human endothelial cell proliferation and survival in vitro reveal a selective antiangiogenic window for various chemotherapeutic drugs. Cancer Res, 2002. 62(23): p. 6938-43.

212. Klement, G., et al., Differences in therapeutic indexes of combination metronomic chemotherapy and an anti-VEGFR-2 antibody in multidrug-resistant human breast cancer xenografts. Clin Cancer Res, 2002. 8(1): p. 221-32.

213. Wang, J., et al., Paclitaxel at ultra low concentrations inhibits angiogenesis without affecting cellular microtubule assembly. Anticancer Drugs, 2003. 14(1): p. 13-9.

214. Vacca, A., et al., Docetaxel versus paclitaxel for antiangiogenesis. J Hematother Stem Cell Res, 2002. 11(1): p. 103-18.

215. Easty, D.M., et al., 10 Human Carcinoma Cell-Lines Derived from Squamous Carcinomas of the Head and Neck. British Journal of Cancer, 1981. 43(6): p. 772-785.

216. Ellis, E.L. and M. Delbruck, The Growth of Bacteriophage. J Gen Physiol, 1939. 22(3): p. 365-84.

217. Chou, T., Relationships between inhibition constants and fractional inhibition in enzyme-catalyzed reactions with different numbers of reactants, different reaction mechanisms, and different types and mechanisms of inhibition. Mol Pharmacol, 1974. 10(2): p. 235-47.

218. Livak, K.J. and T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 2001. 25(4): p. 402-8.

219. Wang, Y., et al., Overexpression of yes-associated protein contributes to progression and poor prognosis of non-small-cell lung cancer. Cancer Science, 2010. 101(5): p. 1279-1285.

220. Yu, F.X., et al., Regulation of the Hippo-YAP Pathway by G-Protein-Coupled Receptor Signaling. Cell, 2012. 150(4): p. 780-791.

221. Hierholzer, J.C.a.K., R.A, Virology Methods Manual. 1996. 222. Zhang, L., et al., The hippo pathway effector YAP regulates motility, invasion, and castration-

resistant growth of prostate cancer cells. Mol Cell Biol, 2015. 35(8): p. 1350-62. 223. Zhao, W., et al., A New Bliss Independence Model to Analyze Drug Combination Data. J

Biomol Screen, 2014. 19(5): p. 817-821. 224. Bliss, C.I., The calculation of microbial assays. Bacteriol Rev, 1956. 20(4): p. 243-58.

258

225. Adair, R.A., et al., Cytotoxic and immune-mediated killing of human colorectal cancer by reovirus-loaded blood and liver mononuclear cells. Int J Cancer, 2013. 132(10): p. 2327-38.

226. Maitra, R., et al., Oncolytic reovirus preferentially induces apoptosis in KRAS mutant colorectal cancer cells, and synergizes with irinotecan. Oncotarget, 2014. 5(9): p. 2807-19.

227. Parrish, C., et al., Oncolytic reovirus enhances rituximab-mediated antibody-dependent cellular cytotoxicity against chronic lymphocytic leukaemia. Leukemia, 2015. 29(9): p. 1799-810.

228. Carew, J.S., et al., Reolysin is a novel reovirus-based agent that induces endoplasmic reticular stress-mediated apoptosis in pancreatic cancer. Cell Death Dis, 2013. 4: p. e728.

229. Pan, D., et al., Stabilisation of p53 enhances reovirus-induced apoptosis and virus spread through p53-dependent NF-kappaB activation. Br J Cancer, 2011. 105(7): p. 1012-22.

230. Anafu, A.A., et al., Interferon-inducible transmembrane protein 3 (IFITM3) restricts reovirus cell entry. J Biol Chem, 2013. 288(24): p. 17261-71.

231. Jennings, V.A., et al., Lymphokine-activated killer and dendritic cell carriage enhances oncolytic reovirus therapy for ovarian cancer by overcoming antibody neutralization in ascites. Int J Cancer, 2014. 134(5): p. 1091-101.

232. Vici, P., et al., Topographic expression of the Hippo transducers TAZ and YAP in triple-negative breast cancer treated with neoadjuvant chemotherapy. J Exp Clin Cancer Res, 2016. 35(1): p. 62.

233. Gurda, G.T., et al., The use of Yes-associated protein expression in the diagnosis of persistent neonatal cholestatic liver disease. Human Pathology, 2014. 45(5): p. 1057-1064.

234. Chou, T.C. and P. Talalay, Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul, 1984. 22: p. 27-55.

235. Ikeda, Y., et al., Reovirus oncolysis in human head and neck squamous carcinoma cells. Auris Nasus Larynx, 2004. 31(4): p. 407-12.

236. Cooper, T., et al., Oncolytic activity of reovirus in HPV positive and negative head and neck squamous cell carcinoma. J Otolaryngol Head Neck Surg, 2015. 44: p. 8.

237. Biotech, O. Clinical Trials. 2016. 238. Prior, I.A., P.D. Lewis, and C. Mattos, A comprehensive survey of Ras mutations in cancer.

Cancer Res, 2012. 72(10): p. 2457-67. 239. P, O.C., P. Rhys-Evans, and S. Eccles, Characterization of ten newly-derived human head and

neck squamous carcinoma cell lines with special reference to c-erbB proto-oncogene expression. Anticancer Res, 2001. 21(3B): p. 1953-63.

240. Berard, A.R., A. Severini, and K.M. Coombs, Comparative proteomic analyses of two reovirus T3D subtypes and comparison to T1L identifies multiple novel proteins in key cellular pathogenic pathways. Proteomics, 2015. 15(12): p. 2113-35.

241. Segal, E., et al., From signatures to models: understanding cancer using microarrays. Nat Genet, 2005. 37 Suppl: p. S38-45.

242. Kominsky, D.J., R.J. Bickel, and K.L. Tyler, Reovirus-induced apoptosis requires mitochondrial release of Smac/DIABLO and involves reduction of cellular inhibitor of apoptosis protein levels. J Virol, 2002. 76(22): p. 11414-24.

243. Clarke, P., et al., Mechanisms of apoptosis during reovirus infection. Curr Top Microbiol Immunol, 2005. 289: p. 1-24.

244. Powell, J., et al., Novel mechanism of impaired function of organic anion-transporting polypeptide 1B3 in human hepatocytes: post-translational regulation of OATP1B3 by protein kinase C activation. Drug Metab Dispos, 2014. 42(11): p. 1964-70.

245. Shibasaki, Y., et al., Expression of indocyanine green-related transporters in hepatocellular carcinoma. J Surg Res, 2015. 193(2): p. 567-76.

246. van de Steeg, E., et al., Complete OATP1B1 and OATP1B3 deficiency causes human Rotor syndrome by interrupting conjugated bilirubin reuptake into the liver. J Clin Invest, 2012. 122(2): p. 519-28.

259

247. Walter, T., et al., O6-Methylguanine-DNA methyltransferase status in neuroendocrine tumours: prognostic relevance and association with response to alkylating agents. Br J Cancer, 2015. 112(3): p. 523-31.

248. Schioth, H.B., et al., Evolutionary origin of amino acid transporter families SLC32, SLC36 and SLC38 and physiological, pathological and therapeutic aspects. Mol Aspects Med, 2013. 34(2-3): p. 571-85.

249. Heublein, S., et al., Proton-assisted amino-acid transporters are conserved regulators of proliferation and amino-acid-dependent mTORC1 activation. Oncogene, 2010. 29(28): p. 4068-79.

250. Boggiano, J.C. and R.G. Fehon, Growth control by committee: intercellular junctions, cell polarity, and the cytoskeleton regulate Hippo signaling. Dev Cell, 2012. 22(4): p. 695-702.

251. Harvey, K.F., X.M. Zhang, and D.M. Thomas, The Hippo pathway and human cancer. Nature Reviews Cancer, 2013. 13(4): p. 246-257.

252. Zhao, B., et al., Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth control. Genes Dev, 2007. 21(21): p. 2747-61.

253. Zhao, B., et al., The Hippo-YAP pathway in organ size control and tumorigenesis: an updated version. Genes & Development, 2010. 24(9): p. 862-874.

254. Steinhardt, A.A., et al., Expression of Yes-associated protein in common solid tumors. Hum Pathol, 2008. 39(11): p. 1582-9.

255. Zender, L., et al., Identification and validation of oncogenes in liver cancer using an integrative oncogenomic approach. Cell, 2006. 125(7): p. 1253-1267.

256. Dong, J.X., et al., Elucidation of a universal size-control mechanism in Drosophila and mammals. Cell, 2007. 130(6): p. 1120-1133.

257. Demirkan, A., et al., Genome-wide association study identifies novel loci associated with circulating phospho- and sphingolipid concentrations. PLoS Genet, 2012. 8(2): p. e1002490.

258. Hao, Y., et al., P2Y6 receptor-mediated proinflammatory signaling in human bronchial epithelia. PLoS One, 2014. 9(9): p. e106235.

259. Brinson, A.E. and T.K. Harden, Differential regulation of the uridine nucleotide-activated P2Y4 and P2Y6 receptors. SER-333 and SER-334 in the carboxyl terminus are involved in agonist-dependent phosphorylation desensitization and internalization of the P2Y4 receptor. J Biol Chem, 2001. 276(15): p. 11939-48.

260. Uren, A.G., et al., Cloning and expression of apoptosis inhibitory protein homologs that function to inhibit apoptosis and/or bind tumor necrosis factor receptor-associated factors. Proc Natl Acad Sci U S A, 1996. 93(10): p. 4974-8.

261. Roy, N., et al., The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J, 1997. 16(23): p. 6914-25.

262. Bertrand, M.J.M., et al., cIAP1 and cIAP2 facilitate cancer cell survival by functioning as E3 ligases that promote RIP1 ubiquitination. Molecular Cell, 2008. 30(6): p. 689-700.

263. Wolin, S.L. and T. Cedervall, The La protein. Annu Rev Biochem, 2002. 71: p. 375-403. 264. Gao, T., Y. Nie, and J. Guo, Hypermethylation of the gene LARP2 for noninvasive prenatal

diagnosis of beta-thalassemia based on DNA methylation profile. Mol Biol Rep, 2012. 39(6): p. 6591-8.

265. Geraghty, R.J., et al., Guidelines for the use of cell lines in biomedical research. British Journal of Cancer, 2014. 111(6): p. 1021-1046.

266. Oka, T., et al., Functional complexes between YAP2 and ZO-2 are PDZ domain-dependent, and regulate YAP2 nuclear localization and signalling. Biochem J, 2010. 432(3): p. 461-72.

267. Bazzoni, G., The JAM family of junctional adhesion molecules. Curr Opin Cell Biol, 2003. 15(5): p. 525-30.

268. Hamazaki, Y., et al., Multi-PDZ domain protein 1 (MUPP1) is concentrated at tight junctions through its possible interaction with claudin-1 and junctional adhesion molecule. J Biol Chem, 2002. 277(1): p. 455-61.

260

269. Snijders, A.M., et al., Rare amplicons implicate frequent deregulation of cell fate specification pathways in oral squamous cell carcinoma. Oncogene, 2005. 24(26): p. 4232-42.

270. Baldwin, C., et al., Multiple microalterations detected at high frequency in oral cancer. Cancer Res, 2005. 65(17): p. 7561-7.

271. Zhang, X., et al., The Hippo pathway transcriptional co-activator, YAP, is an ovarian cancer oncogene. Oncogene, 2011. 30(25): p. 2810-22.

272. Konsavage, W.M., Jr. and G.S. Yochum, Intersection of Hippo/YAP and Wnt/beta-catenin signaling pathways. Acta Biochim Biophys Sin (Shanghai), 2013. 45(2): p. 71-9.

273. Bao, Y.J., et al., A cell-based assay to screen stimulators of the Hippo pathway reveals the inhibitory effect of dobutamine on the YAP-dependent gene transcription. Journal of Biochemistry, 2011. 150(2): p. 199-208.

274. Zhao, B., et al., A coordinated phosphorylation by Lats and CK1 regulates YAP stability through SCF(beta-TRCP). Genes Dev, 2010. 24(1): p. 72-85.

275. Piccolo, S., M. Cordenonsi, and S. Dupont, Molecular pathways: YAP and TAZ take center stage in organ growth and tumorigenesis. Clin Cancer Res, 2013. 19(18): p. 4925-30.

276. Piccolo, S., S. Dupont, and M. Cordenonsi, The biology of YAP/TAZ: hippo signaling and beyond. Physiol Rev, 2014. 94(4): p. 1287-312.

277. Sudol, M., D.C. Shields, and A. Farooq, Structures of YAP protein domains reveal promising targets for development of new cancer drugs. Seminars in Cell & Developmental Biology, 2012. 23(7): p. 827-833.

278. Kim, M. and E.H. Jho, Cross-talk between Wnt/beta-catenin and Hippo signaling pathways: a brief review. Bmb Reports, 2014. 47(10): p. 540-545.

279. Hisaoka, M., A. Tanaka, and H. Hashimoto, Molecular alterations of h-warts/LATS1 tumor suppressor in human soft tissue sarcoma. Lab Invest, 2002. 82(10): p. 1427-35.

280. Chakraborty, S., et al., Identification of genes associated with tumorigenesis of retinoblastoma by microarray analysis. Genomics, 2007. 90(3): p. 344-53.

281. Jimenez-Velasco, A., et al., Downregulation of the large tumor suppressor 2 (LATS2/KPM) gene is associated with poor prognosis in acute lymphoblastic leukemia. Leukemia, 2005. 19(12): p. 2347-50.

282. Zhou, D., et al., Mst1 and Mst2 maintain hepatocyte quiescence and suppress hepatocellular carcinoma development through inactivation of the Yap1 oncogene. Cancer Cell, 2009. 16(5): p. 425-38.

283. Xu, M.Z., et al., Yes-associated protein is an independent prognostic marker in hepatocellular carcinoma. Cancer, 2009. 115(19): p. 4576-85.

284. Banks, L., D. Pim, and M. Thomas, Human tumour viruses and the deregulation of cell polarity in cancer. Nat Rev Cancer, 2012. 12(12): p. 877-86.

285. Zhang, T., et al., Hepatitis B Virus X Protein Modulates Oncogene Yes-Associated Protein by CREB to Promote Growth of Hepatoma Cells. Hepatology, 2012. 56(6): p. 2051-2059.

286. Bonilla-Delgado, J., et al., The E6 Oncoprotein from HPV16 Enhances the Canonical Wnt/beta-Catenin Pathway in Skin Epidermis In Vivo. Molecular Cancer Research, 2012. 10(2): p. 250-258.

287. Varelas, X., et al., The Crumbs Complex Couples Cell Density Sensing to Hippo-Dependent Control of the TGF-beta-SMAD Pathway. Developmental Cell, 2010. 19(6): p. 831-844.

288. Tsai, C.N., et al., The Epstein-Barr virus oncogene product, latent membrane protein 1, induces the down-regulation of E-cadherin gene expression via activation of DNA methyltransferases. Proceedings of the National Academy of Sciences of the United States of America, 2002. 99(15): p. 10084-10089.

289. Morrison, J.A. and N. Raab-Traub, Roles of the ITAM and PY motifs of Epstein-Barr virus latent membrane protein 2A in the inhibition of epithelial cell differentiation and activation of beta-catenin signaling. Journal of Virology, 2005. 79(4): p. 2375-2382.

261

290. Ma, G., et al., HTLV-1 bZIP factor dysregulates the Wnt pathways to support proliferation and migration of adult T-cell leukemia cells. Oncogene, 2013. 32(36): p. 4222-4230.

291. Liu, G., et al., Kaposi sarcoma-associated herpesvirus promotes tumorigenesis by modulating the Hippo pathway. Oncogene, 2015. 34(27): p. 3536-46.

292. Kim, T.K. and J.H. Eberwine, Mammalian cell transfection: the present and the future. Anal Bioanal Chem, 2010. 397(8): p. 3173-8.

293. Xu, M.Z., et al., AXL receptor kinase is a mediator of YAP-dependent oncogenic functions in hepatocellular carcinoma. Oncogene, 2011. 30(10): p. 1229-1240.

294. Tsujiura, M., et al., Yes-Associated Protein (YAP) Modulates Oncogenic Features and Radiation Sensitivity in Endometrial Cancer. Plos One, 2014. 9(6).

295. Huang, J., et al., The Hippo signaling pathway coordinately regulates cell proliferation and apoptosis by inactivating Yorkie, the Drosophila Homolog of YAP. Cell, 2005. 122(3): p. 421-34.

296. Santos, S.D., P.J. Verveer, and P.I. Bastiaens, Growth factor-induced MAPK network topology shapes Erk response determining PC-12 cell fate. Nat Cell Biol, 2007. 9(3): p. 324-30.

297. Mo, J.S., et al., Regulation of the Hippo-YAP pathway by protease-activated receptors (PARs). Genes Dev, 2012. 26(19): p. 2138-43.

298. Liu-Chittenden, Y., et al., Genetic and pharmacological disruption of the TEAD-YAP complex suppresses the oncogenic activity of YAP. Genes Dev, 2012. 26(12): p. 1300-5.

299. Sansores-Garcia, L., et al., Modulating F-actin organization induces organ growth by affecting the Hippo pathway. EMBO J, 2011. 30(12): p. 2325-35.

300. Rosenbluh, J., et al., beta-Catenin-driven cancers require a YAP1 transcriptional complex for survival and tumorigenesis. Cell, 2012. 151(7): p. 1457-73.

301. Basu, S., et al., Akt phosphorylates the Yes-associated protein, YAP, to induce interaction with 14-3-3 and attenuation of p73-mediated apoptosis. Mol Cell, 2003. 11(1): p. 11-23.

302. Azzolin, L., et al., YAP/TAZ Incorporation in the beta-Catenin Destruction Complex Orchestrates the Wnt Response. Cell, 2014. 158(1): p. 157-170.

303. Fujii, M., et al., TGF-beta synergizes with defects in the Hippo pathway to stimulate human malignant mesothelioma growth. J Exp Med, 2012. 209(3): p. 479-94.

304. Fernandez, A., et al., YAP1 is amplified and up-regulated in hedgehog-associated medulloblastomas and mediates Sonic hedgehog-driven neural precursor proliferation. Genes & Development, 2009. 23(23): p. 2729-2741.

305. Seo, Y.J., et al., Sphingosine 1-phosphate-metabolizing enzymes control influenza virus propagation and viral cytopathogenicity. J Virol, 2010. 84(16): p. 8124-31.

306. Sharma, N., A.S. Akhade, and A. Qadri, Sphingosine-1-phosphate suppresses TLR-induced CXCL8 secretion from human T cells. Journal of Leukocyte Biology, 2013. 93(4): p. 521-528.

307. Pyne, N.J. and S. Pyne, Sphingosine 1-phosphate and cancer. Nat Rev Cancer, 2010. 10(7): p. 489-503.

308. Coschi, C.H. and F.A. Dick, Chromosome instability and deregulated proliferation: an unavoidable duo. Cell Mol Life Sci, 2012. 69(12): p. 2009-24.

309. Kim, M., et al., Acquired resistance to reoviral oncolysis in Ras-transformed fibrosarcoma cells. Oncogene, 2007. 26(28): p. 4124-34.

310. Kim, M., et al., The viral tropism of two distinct oncolytic viruses, reovirus and myxoma virus, is modulated by cellular tumor suppressor gene status. Oncogene, 2010. 29(27): p. 3990-6.

311. Kim, M., Naturally occurring reoviruses for human cancer therapy. BMB Rep, 2015. 48(8): p. 454-60.

312. Bischoff, J.R., et al., An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. Science, 1996. 274(5286): p. 373-6.

313. Heinemann, L., et al., The effect of cell cycle synchronization on tumor sensitivity to reovirus oncolysis. Mol Ther, 2010. 18(12): p. 2085-93.

262

314. Berger, A.K. and P. Danthi, Reovirus activates a caspase-independent cell death pathway. MBio, 2013. 4(3): p. e00178-13.

315. Gong, J. and M.M. Mita, Activated ras signaling pathways and reovirus oncolysis: an update on the mechanism of preferential reovirus replication in cancer cells. Front Oncol, 2014. 4: p. 167.

316. Wu, W., P. Liu, and J. Li, Necroptosis: an emerging form of programmed cell death. Crit Rev Oncol Hematol, 2012. 82(3): p. 249-58.

317. Vandenabeele, P., et al., Molecular mechanisms of necroptosis: an ordered cellular explosion. Nat Rev Mol Cell Biol, 2010. 11(10): p. 700-14.

318. Thirukkumaran, C.M., et al., Reovirus modulates autophagy during oncolysis of multiple myeloma. Autophagy, 2013. 9(3): p. 413-4.

319. Qin, L., et al., ER stress negatively regulates AKT/TSC/mTOR pathway to enhance autophagy. Autophagy, 2010. 6(2): p. 239-47.

320. Meng, S., et al., Avian reovirus triggers autophagy in primary chicken fibroblast cells and Vero cells to promote virus production. Arch Virol, 2012. 157(4): p. 661-8.

321. Connolly, J.L. and T.S. Dermody, Virion disassembly is required for apoptosis induced by reovirus. J Virol, 2002. 76(4): p. 1632-41.

322. Coffey, C.M., et al., Reovirus outer capsid protein micro1 induces apoptosis and associates with lipid droplets, endoplasmic reticulum, and mitochondria. J Virol, 2006. 80(17): p. 8422-38.

323. Kominsky, D.J., R.J. Bickel, and K.L. Tyler, Reovirus-induced apoptosis requires both death receptor- and mitochondrial-mediated caspase-dependent pathways of cell death. Cell Death Differ, 2002. 9(9): p. 926-33.

324. Clarke, P., et al., Reovirus-induced apoptosis is mediated by TRAIL. J Virol, 2000. 74(17): p. 8135-9.

325. Clarke, P., et al., Caspase 8-dependent sensitization of cancer cells to TRAIL-induced apoptosis following reovirus-infection. Oncogene, 2001. 20(47): p. 6910-9.

326. Knowlton, J.J., T.S. Dermody, and G.H. Holm, Apoptosis induced by mammalian reovirus is beta interferon (IFN) independent and enhanced by IFN regulatory factor 3- and NF-kappaB-dependent expression of Noxa. J Virol, 2012. 86(3): p. 1650-60.

327. Clarke, P., et al., Reovirus infection activates JNK and the JNK-dependent transcription factor c-Jun. J Virol, 2001. 75(23): p. 11275-83.

328. Oslowski, C.M. and F. Urano, Measuring ER stress and the unfolded protein response using mammalian tissue culture system. Methods Enzymol, 2011. 490: p. 71-92.

329. Kelly, K.R., et al., Reovirus therapy stimulates endoplasmic reticular stress, NOXA induction, and augments bortezomib-mediated apoptosis in multiple myeloma. Oncogene, 2012. 31(25): p. 3023-38.

330. Roulstone, V., et al., BRAF- and MEK-Targeted Small Molecule Inhibitors Exert Enhanced Antimelanoma Effects in Combination With Oncolytic Reovirus Through ER Stress. Mol Ther, 2015. 23(5): p. 931-42.

331. Randall, R.E. and S. Goodbourn, Interferons and viruses: an interplay between induction, signalling, antiviral responses and virus countermeasures. Journal of General Virology, 2008. 89: p. 1-47.

332. Sherry, B., Rotavirus and reovirus modulation of the interferon response. J Interferon Cytokine Res, 2009. 29(9): p. 559-67.

333. Shmulevitz, M., et al., Oncogenic Ras promotes reovirus spread by suppressing IFN-beta production through negative regulation of RIG-I signaling. Cancer Res, 2010. 70(12): p. 4912-21.

334. Rudd, P. and G. Lemay, Correlation between interferon sensitivity of reovirus isolates and ability to discriminate between normal and Ras-transformed cells. J Gen Virol, 2005. 86(Pt 5): p. 1489-97.

263

335. Errington, F., et al., Inflammatory tumour cell killing by oncolytic reovirus for the treatment of melanoma. Gene Ther, 2008. 15(18): p. 1257-70.

336. Roulstone, V., et al., Phase I trial of cyclophosphamide as an immune modulator for optimizing oncolytic reovirus delivery to solid tumors. Clin Cancer Res, 2015. 21(6): p. 1305-12.

337. Matsui, Y. and Z.C. Lai, Mutual regulation between Hippo signaling and actin cytoskeleton. Protein Cell, 2013. 4(12): p. 904-10.

338. Coffey, M.C., et al., Reovirus therapy of tumors with activated Ras pathway. Science, 1998. 282(5392): p. 1332-4.

339. Marcato, P., et al., Ras transformation mediates reovirus oncolysis by enhancing virus uncoating, particle infectivity, and apoptosis-dependent release. Mol Ther, 2007. 15(8): p. 1522-30.

340. Smakman, N., et al., Sensitization to apoptosis underlies KrasD12-dependent oncolysis of murine C26 colorectal carcinoma cells by reovirus T3D. J Virol, 2005. 79(23): p. 14981-5.

341. Atlas, T.H.P. YAP1 Prostate Cancer. 2016 10-05-16]; Available from: http://www.proteinatlas.org/ENSG00000137693-YAP1/cancer/tissue/prostate+cancer.

342. Zhao, B., et al., Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth control. Genes & Development, 2007. 21(21): p. 2747-2761.

343. Berard, A.R., A. Severini, and K.M. Coombs, Differential Reovirus-Specific and Herpesvirus-Specific Activator Protein 1 Activation of Secretogranin II Leads to Altered Virus Secretion. J Virol, 2015. 89(23): p. 11954-64.

344. Poggioli, G.J., et al., Reovirus-induced alterations in gene expression related to cell cycle regulation. J Virol, 2002. 76(6): p. 2585-94.

345. Clarke, P. and K.L. Tyler, Down-regulation of cFLIP following reovirus infection sensitizes human ovarian cancer cells to TRAIL-induced apoptosis. Apoptosis, 2007. 12(1): p. 211-23.

346. Cho, I.R., et al., Reovirus infection induces apoptosis of TRAIL-resistant gastric cancer cells by down-regulation of Akt activation. Int J Oncol, 2010. 36(4): p. 1023-30.

347. Mohamed, A., R.N. Johnston, and M. Shmulevitz, Potential for Improving Potency and Specificity of Reovirus Oncolysis with Next-Generation Reovirus Variants. Viruses, 2015. 7(12): p. 6251-78.

348. van den Hengel, S.K., et al., Heterogeneous reovirus susceptibility in human glioblastoma stem-like cell cultures. Cancer Gene Ther, 2013. 20(9): p. 507-13.

349. van den Wollenberg, D.J., et al., Isolation of reovirus T3D mutants capable of infecting human tumor cells independent of junction adhesion molecule-A. PLoS One, 2012. 7(10): p. e48064.

350. Smakman, N., et al., KRAS(D13) Promotes apoptosis of human colorectal tumor cells by ReovirusT3D and oxaliplatin but not by tumor necrosis factor-related apoptosis-inducing ligand. Cancer Res, 2006. 66(10): p. 5403-8.

351. Tyler, K.L., et al., Differences in the capacity of reovirus strains to induce apoptosis are determined by the viral attachment protein sigma 1. J Virol, 1995. 69(11): p. 6972-9.

352. Rodgers, S.E., et al., Reovirus-induced apoptosis of MDCK cells is not linked to viral yield and is blocked by Bcl-2. J Virol, 1997. 71(3): p. 2540-6.

353. Imani, F. and B.L. Jacobs, Inhibitory activity for the interferon-induced protein kinase is associated with the reovirus serotype 1 sigma 3 protein. Proc Natl Acad Sci U S A, 1988. 85(21): p. 7887-91.

354. Forbes, N.E., et al., Exploiting tumor epigenetics to improve oncolytic virotherapy. Front Genet, 2013. 4: p. 184.

355. Kirn, D.H., et al., Targeting of interferon-beta to produce a specific, multi-mechanistic oncolytic vaccinia virus. PLoS Med, 2007. 4(12): p. e353.

http://www.proteinatlas.org/ENSG00000137693-YAP1/cancer/tissue/prostate+cancer

264

356. Stojdl, D.F., et al., VSV strains with defects in their ability to shutdown innate immunity are potent systemic anti-cancer agents. Cancer Cell, 2003. 4(4): p. 263-75.

357. Lin, L., et al., The Hippo effector YAP promotes resistance to RAF- and MEK-targeted cancer therapies. Nat Genet, 2015. 47(3): p. 250-6.

358. UK, C.R. Prostate cancer incidence statistics. 2014 07/05/14; Available from: http://www.cancerresearchuk.org/cancer-info/cancerstats/types/prostate/incidence/.

359. Mainou, B.A., et al., Reovirus cell entry requires functional microtubules. MBio, 2013. 4(4). 360. Bennett, N.C., et al., Expression profiles and functional associations of endogenous androgen

receptor and caveolin-1 in prostate cancer cell lines. Prostate, 2014. 74(5): p. 478-87. 361. Takahashi, T., et al., The study of PSA gene expression on urogenital cell lines. Int J Urol,

1999. 6(10): p. 526-31. 362. Zhu, M.L., et al., Tubulin-targeting chemotherapy impairs androgen receptor activity in

prostate cancer. Cancer Res, 2010. 70(20): p. 7992-8002. 363. Eisenberg, D.P., et al., 5-fluorouracil and gemcitabine potentiate the efficacy of oncolytic

herpes viral gene therapy in the treatment of pancreatic cancer. J Gastrointest Surg, 2005. 9(8): p. 1068-77; discussion 1077-9.

364. Adusumilli, P.S., et al., Cisplatin-induced GADD34 upregulation potentiates oncolytic viral therapy in the treatment of malignant pleural mesothelioma. Cancer Biol Ther, 2006. 5(1): p. 48-53.

365. Petrowsky, H., et al., Functional interaction between fluorodeoxyuridine-induced cellular alterations and replication of a ribonucleotide reductase-negative herpes simplex virus. J Virol, 2001. 75(15): p. 7050-8.

366. Gutermann, A., et al., Efficacy of oncolytic herpesvirus NV1020 can be enhanced by combination with chemotherapeutics in colon carcinoma cells. Hum Gene Ther, 2006. 17(12): p. 1241-53.

367. Pawlik, T.M., et al., Prodrug bioactivation and oncolysis of diffuse liver metastases by a herpes simplex virus 1 mutant that expresses the CYP2B1 transgene. Cancer, 2002. 95(5): p. 1171-81.

368. Raki, M., et al., Combination of gemcitabine and Ad5/3-Delta24, a tropism modified conditionally replicating adenovirus, for the treatment of ovarian cancer. Gene Ther, 2005. 12(15): p. 1198-205.

369. Tyminski, E., et al., Brain tumor oncolysis with replication-conditional herpes simplex virus type 1 expressing the prodrug-activating genes, CYP2B1 and secreted human intestinal carboxylesterase, in combination with cyclophosphamide and irinotecan. Cancer Res, 2005. 65(15): p. 6850-7.

370. Donaldson, K.L., et al., Activation of p34cdc2 coincident with taxol-induced apoptosis. Cell Growth Differ, 1994. 5(10): p. 1041-50.

371. Strobel, T., et al., BAX enhances paclitaxel-induced apoptosis through a p53-independent pathway. Proc Natl Acad Sci U S A, 1996. 93(24): p. 14094-9.

372. Haldar, S., J. Chintapalli, and C.M. Croce, Taxol induces bcl-2 phosphorylation and death of prostate cancer cells. Cancer Res, 1996. 56(6): p. 1253-5.

373. Corthay, A., How do Regulatory T Cells Work? Scandinavian Journal of Immunology, 2009. 70(4): p. 326-336.

374. Adeegbe, D.O. and H. Nishikawa, Natural and induced T regulatory cells in cancer. Frontiers in Immunology, 2013. 4.

375. Tagliamonte, M., et al., A novel multi-drug metronomic chemotherapy significantly delays tumor growth in mice. J Transl Med, 2016. 14(1): p. 58.

376. Rajani, K., et al., Combination Therapy With Reovirus and Anti-PD-1 Blockade Controls Tumor Growth Through Innate and Adaptive Immune Responses. Mol Ther, 2016. 24(1): p. 166-74.

377. Ge, L., et al., Yes-associated protein expression in head and neck squamous cell carcinoma nodal metastasis. PLoS One, 2011. 6(11): p. e27529.

http://www.cancerresearchuk.org/cancer-info/cancerstats/types/prostate/incidence/

265

378. Xu, C.M., et al., Mst1 overexpression inhibited the growth of human non-small cell lung cancer in vitro and in vivo. Cancer Gene Ther, 2013. 20(8): p. 453-60.

379. Konsavage, W.M., Jr., et al., Wnt/beta-catenin signaling regulates Yes-associated protein (YAP) gene expression in colorectal carcinoma cells. J Biol Chem, 2012. 287(15): p. 11730-9.

380. Wang, C., et al., Differences in Yes-associated protein and mRNA levels in regenerating liver and hepatocellular carcinoma. Mol Med Rep, 2012. 5(2): p. 410-4.

381. Kang, W., et al., Yes-associated protein 1 exhibits oncogenic property in gastric cancer and its nuclear accumulation associates with poor prognosis. Clin Cancer Res, 2011. 17(8): p. 2130-9.

382. Dunn, G.P., C.M. Koebel, and R.D. Schreiber, Interferons, immunity and cancer immunoediting. Nat Rev Immunol, 2006. 6(11): p. 836-48.

383. Reddy, P., et al., Actin cytoskeleton regulates Hippo signaling. PLoS One, 2013. 8(9): p. e73763.

384. Fernandez, B.G., et al., Actin-Capping Protein and the Hippo pathway regulate F-actin and tissue growth in Drosophila. Development, 2011. 138(11): p. 2337-46.

385. Visser-Grieve, S., et al., LATS1 tumor suppressor is a novel actin-binding protein and negative regulator of actin polymerization. Cell Res, 2011. 21(10): p. 1513-6.

386. Kuser-Abali, G., et al., YAP1 and AR interactions contribute to the switch from androgen-dependent to castration-resistant growth in prostate cancer. Nature Communications, 2015. 6.

387. Wang, P., et al., PP1A-mediated dephosphorylation positively regulates YAP2 activity. PLoS One, 2011. 6(9): p. e24288.

OPTIMISING REOVIRUS TYPE 3 DEARING (REOLYSIN®) AS AN …epubs.surrey.ac.uk/811663/1/FINAL THESIS...

Documents

Transcript of OPTIMISING REOVIRUS TYPE 3 DEARING (REOLYSIN®) AS AN …epubs.surrey.ac.uk/811663/1/FINAL THESIS...